OpenwaterWiki mediawiki https://wiki.openwater.health/index.php/Main_Page MediaWiki 1.40.1 first-letter Media Special Talk User User talk OpenwaterWiki OpenwaterWiki talk File File talk MediaWiki MediaWiki talk Template Template talk Help Help talk Category Category talk 0 75 338 2023-12-14T00:26:45Z Opw12 8 Opw12 moved page [[Acousto Optic]] to [[Holographic Acousto Optic Imaging]] 338 wikitext text/x-wiki #REDIRECT [[Holographic Acousto Optic Imaging]] 1wp7tgr5bmcs4m37su97txh0o9ra3hw 0 4 4 2023-12-12T20:16:51Z OpenwaterAndrew 3 Created page with "Please visit the opw_bloodflow_gen1_hw GitHub repository for the complete collection of files referenced on this page. = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. Features - Measure and display blood flow inde..." 4 wikitext text/x-wiki Please visit the opw_bloodflow_gen1_hw GitHub repository for the complete collection of files referenced on this page. = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. Features - Measure and display blood flow index and blood volume index - Included a hand held wand, so many places on the scalp could be measured to identify the most important areas with data - Data processed on the embedded computer - Movable cart for hospital environment = Gen 1 to Gen 1B Device History = The first device sent to a hospital was called a Blood flow Gen 1 cart, This device was approved by the hospital principal investigator and IRB. After use in the hospital setting we received input that the Gen 1 device was too hard to move into the small rooms needed for patient measurement. The Gen 1B device was developed to be a are-configured device that would take up less floor space, in addition the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that would mount to an IV pole. The Gen 1B device yielded important information on the use with a clinically significant population and operated by research coordinators or nurse practitioners in a hospital. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is more completely detailed and is essentially an improved version. The Gen 1B device was made up of large off the shelf general components, whereas the Gen 2 device used more customized components that were smaller and less expensive to make. The data produced from the Gen 1B is included in the blood flow Gen 1 analysis repo. One of the most significant hardware results from Gen 1B involved finding design criteria to move from a wand based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurement. With the wand we could interrogate everywhere on the scalp, with a headset we would only look at a few positions. Source fiber with Diffusing Spacer on Gen 1B Wand For the highest percentage of patients, the best data was found to come from areas on the scalp with no or few hair follicles (or the forehead area). A second study was launched to look at 32 possible positions on the forehead which included left and right hemispheres. The results of the forehead study informed a design for the headset used in Gen 2. See the opw_bloodflow_gen2_hw GitHub repository for Gen 2 device designs. Some of the detail for the Gen 1B device is omitted because there are improved versions of the entire device and subassemblies readily available in the Gen 2 repo. The Gen 2 device was fully documented and was designed as the manufacturable replacement for the Gen 1B prototype device. 03dy1wnktlbvzzszc6fwl0gyi79o2rh 5 4 2023-12-12T21:19:33Z Openwaterpete 5 /* Gen 1 to Gen 1B Device History */ 5 wikitext text/x-wiki Please visit the opw_bloodflow_gen1_hw GitHub repository for the complete collection of files referenced on this page. = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. Features - Measure and display blood flow index and blood volume index - Included a hand held wand, so many places on the scalp could be measured to identify the most important areas with data - Data processed on the embedded computer - Movable cart for hospital environment = Gen 1 to Gen 1B Device History = The first device sent to a hospital was called a Blood flow Gen 1 cart, This device was approved by the hospital principal investigator and IRB. After use in the hospital setting we received input that the Gen 1 device was too hard to move into the small rooms needed for patient measurement. The Gen 1B device was developed to be a are-configured device that would take up less floor space, in addition the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that would mount to an IV pole. The Gen 1B device yielded important information on the use with a clinically significant population and operated by research coordinators or nurse practitioners in a hospital. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is more completely detailed and is essentially an improved version. The Gen 1B device was made up of large off the shelf general components, whereas the Gen 2 device used more customized components that were smaller and less expensive to make. The data produced from the Gen 1B is included in the blood flow Gen 1 analysis repo. One of the most significant hardware results from Gen 1B involved finding design criteria to move from a wand based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurement. With the wand we could interrogate everywhere on the scalp, with a headset we would only look at a few positions. Source fiber with Diffusing Spacer on [file:wand gen1b.jpg] Gen 1B Wand For the highest percentage of patients, the best data was found to come from areas on the scalp with no or few hair follicles (or the forehead area). A second study was launched to look at 32 possible positions on the forehead which included left and right hemispheres. The results of the forehead study informed a design for the headset used in Gen 2. See the opw_bloodflow_gen2_hw GitHub repository for Gen 2 device designs. Some of the detail for the Gen 1B device is omitted because there are improved versions of the entire device and subassemblies readily available in the Gen 2 repo. The Gen 2 device was fully documented and was designed as the manufacturable replacement for the Gen 1B prototype device. s428aj6yjblqg24xu7wx7bdgvmuo8l3 6 5 2023-12-12T21:20:17Z Openwaterpete 5 6 wikitext text/x-wiki Please visit the opw_bloodflow_gen1_hw GitHub repository for the complete collection of files referenced on this page. = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. Features - Measure and display blood flow index and blood volume index - Included a hand held wand, so many places on the scalp could be measured to identify the most important areas with data - Data processed on the embedded computer - Movable cart for hospital environment = Gen 1 to Gen 1B Device History = The first device sent to a hospital was called a Blood flow Gen 1 cart, This device was approved by the hospital principal investigator and IRB. After use in the hospital setting we received input that the Gen 1 device was too hard to move into the small rooms needed for patient measurement. The Gen 1B device was developed to be a are-configured device that would take up less floor space, in addition the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that would mount to an IV pole. The Gen 1B device yielded important information on the use with a clinically significant population and operated by research coordinators or nurse practitioners in a hospital. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is more completely detailed and is essentially an improved version. The Gen 1B device was made up of large off the shelf general components, whereas the Gen 2 device used more customized components that were smaller and less expensive to make. The data produced from the Gen 1B is included in the blood flow Gen 1 analysis repo. One of the most significant hardware results from Gen 1B involved finding design criteria to move from a wand based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurement. With the wand we could interrogate everywhere on the scalp, with a headset we would only look at a few positions. Source fiber with Diffusing Spacer on [file:https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw/blob/main/wand%20gen1b.jpg] Gen 1B Wand For the highest percentage of patients, the best data was found to come from areas on the scalp with no or few hair follicles (or the forehead area). A second study was launched to look at 32 possible positions on the forehead which included left and right hemispheres. The results of the forehead study informed a design for the headset used in Gen 2. See the opw_bloodflow_gen2_hw GitHub repository for Gen 2 device designs. Some of the detail for the Gen 1B device is omitted because there are improved versions of the entire device and subassemblies readily available in the Gen 2 repo. The Gen 2 device was fully documented and was designed as the manufacturable replacement for the Gen 1B prototype device. iduq9coh3d30okdpxgcayesrpqaozhb 7 6 2023-12-12T21:21:17Z Openwaterpete 5 7 wikitext text/x-wiki Please visit the opw_bloodflow_gen1_hw GitHub repository for the complete collection of files referenced on this page. = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. Features - Measure and display blood flow index and blood volume index - Included a hand held wand, so many places on the scalp could be measured to identify the most important areas with data - Data processed on the embedded computer - Movable cart for hospital environment = Gen 1 to Gen 1B Device History = The first device sent to a hospital was called a Blood flow Gen 1 cart, This device was approved by the hospital principal investigator and IRB. After use in the hospital setting we received input that the Gen 1 device was too hard to move into the small rooms needed for patient measurement. The Gen 1B device was developed to be a are-configured device that would take up less floor space, in addition the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that would mount to an IV pole. The Gen 1B device yielded important information on the use with a clinically significant population and operated by research coordinators or nurse practitioners in a hospital. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is more completely detailed and is essentially an improved version. The Gen 1B device was made up of large off the shelf general components, whereas the Gen 2 device used more customized components that were smaller and less expensive to make. The data produced from the Gen 1B is included in the blood flow Gen 1 analysis repo. One of the most significant hardware results from Gen 1B involved finding design criteria to move from a wand based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurement. With the wand we could interrogate everywhere on the scalp, with a headset we would only look at a few positions. Source fiber with Diffusing Spacer on [[file:https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw/blob/main/wand%20gen1b.jpg]] Gen 1B Wand For the highest percentage of patients, the best data was found to come from areas on the scalp with no or few hair follicles (or the forehead area). A second study was launched to look at 32 possible positions on the forehead which included left and right hemispheres. The results of the forehead study informed a design for the headset used in Gen 2. See the opw_bloodflow_gen2_hw GitHub repository for Gen 2 device designs. Some of the detail for the Gen 1B device is omitted because there are improved versions of the entire device and subassemblies readily available in the Gen 2 repo. The Gen 2 device was fully documented and was designed as the manufacturable replacement for the Gen 1B prototype device. m9o7twqqv7ti526p78k6xbpp1qqhake 8 7 2023-12-12T21:22:27Z Openwaterpete 5 8 wikitext text/x-wiki Please visit the opw_bloodflow_gen1_hw GitHub repository for the complete collection of files referenced on this page. = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. Features - Measure and display blood flow index and blood volume index - Included a hand held wand, so many places on the scalp could be measured to identify the most important areas with data - Data processed on the embedded computer - Movable cart for hospital environment = Gen 1 to Gen 1B Device History = The first device sent to a hospital was called a Blood flow Gen 1 cart, This device was approved by the hospital principal investigator and IRB. After use in the hospital setting we received input that the Gen 1 device was too hard to move into the small rooms needed for patient measurement. The Gen 1B device was developed to be a are-configured device that would take up less floor space, in addition the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that would mount to an IV pole. The Gen 1B device yielded important information on the use with a clinically significant population and operated by research coordinators or nurse practitioners in a hospital. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is more completely detailed and is essentially an improved version. The Gen 1B device was made up of large off the shelf general components, whereas the Gen 2 device used more customized components that were smaller and less expensive to make. The data produced from the Gen 1B is included in the blood flow Gen 1 analysis repo. One of the most significant hardware results from Gen 1B involved finding design criteria to move from a wand based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurement. With the wand we could interrogate everywhere on the scalp, with a headset we would only look at a few positions. Source fiber with Diffusing Spacer on: [[file:https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw/blob/main/wand%20gen1b.jpg]] Gen 1B Wand For the highest percentage of patients, the best data was found to come from areas on the scalp with no or few hair follicles (or the forehead area). A second study was launched to look at 32 possible positions on the forehead which included left and right hemispheres. The results of the forehead study informed a design for the headset used in Gen 2. See the opw_bloodflow_gen2_hw GitHub repository for Gen 2 device designs. Some of the detail for the Gen 1B device is omitted because there are improved versions of the entire device and subassemblies readily available in the Gen 2 repo. The Gen 2 device was fully documented and was designed as the manufacturable replacement for the Gen 1B prototype device. ryzzh6i0ayhzzs74hlstqivh77n23vy 12 8 2023-12-12T21:28:08Z Openwaterpete 5 12 wikitext text/x-wiki Please visit the opw_bloodflow_gen1_hw GitHub repository for the complete collection of files referenced on this page. = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. Features - Measure and display blood flow index and blood volume index - Included a hand held wand, so many places on the scalp could be measured to identify the most important areas with data - Data processed on the embedded computer - Movable cart for hospital environment = Gen 1 to Gen 1B Device History = The first device sent to a hospital was called a Blood flow Gen 1 cart, This device was approved by the hospital principal investigator and IRB. After use in the hospital setting we received input that the Gen 1 device was too hard to move into the small rooms needed for patient measurement. The Gen 1B device was developed to be a are-configured device that would take up less floor space, in addition the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that would mount to an IV pole. The Gen 1B device yielded important information on the use with a clinically significant population and operated by research coordinators or nurse practitioners in a hospital. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is more completely detailed and is essentially an improved version. The Gen 1B device was made up of large off the shelf general components, whereas the Gen 2 device used more customized components that were smaller and less expensive to make. The data produced from the Gen 1B is included in the blood flow Gen 1 analysis repo. One of the most significant hardware results from Gen 1B involved finding design criteria to move from a wand based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurement. With the wand we could interrogate everywhere on the scalp, with a headset we would only look at a few positions. Source fiber with Diffusing Spacer on: [file:https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw/blob/main/wand%20gen1b.jpg] Gen 1B Wand For the highest percentage of patients, the best data was found to come from areas on the scalp with no or few hair follicles (or the forehead area). A second study was launched to look at 32 possible positions on the forehead which included left and right hemispheres. The results of the forehead study informed a design for the headset used in Gen 2. See the opw_bloodflow_gen2_hw GitHub repository for Gen 2 device designs. Some of the detail for the Gen 1B device is omitted because there are improved versions of the entire device and subassemblies readily available in the Gen 2 repo. The Gen 2 device was fully documented and was designed as the manufacturable replacement for the Gen 1B prototype device. mibhaemriitu46box0k2nuhd3jpzde7 49 12 2023-12-12T22:32:04Z 50.227.118.138 49 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw opw_bloodflow_gen1_hw GitHub repository] for the complete collection of files referenced on this page.''' = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. == Features == * Measure and display blood flow index and blood volume index * Included a hand-held wand, allowing measurement on various scalp areas to identify crucial data points * Data processed on the embedded computer * Movable cart designed for hospital environments = Gen 1 to Gen 1B Device History = The first device sent to a hospital was called a Blood Flow Gen 1 cart, approved by the hospital principal investigator and IRB. Feedback received indicated that the Gen 1 device was challenging to maneuver into small rooms needed for patient measurement. The Gen 1B device was then developed as a reconfigured device that occupied less floor space. Additionally, the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that could mount to an IV pole. The Gen 1B device provided valuable insights into its use within a clinically significant population and was operated by research coordinators or nurse practitioners in a hospital setting. Refer to Gen 1 Blood Flow White Paper for detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is a more detailed and improved version. Gen 1B used large off-the-shelf general components, while Gen 2 employed smaller, less expensive customized components. The data produced from Gen 1B is included in the blood flow Gen 1 analysis repo. One significant hardware result from Gen 1B involved finding design criteria to shift from a wand-based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurements. With the headset, only a few positions on the scalp were observed. [[File:wand gen1b.jpg|thumb|Gen 1B Wand]] For the majority of patients, the best data came from areas on the scalp with no or few hair follicles (e.g., the forehead area). A study was conducted to examine 32 possible positions on the forehead, informing the design for the headset used in Gen 2. Refer to the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] for Gen 2 device designs. Some details of the Gen 1B device are omitted due to the availability of improved versions in the Gen 2 repository. Gen 2 was fully documented and designed as the manufacturable replacement for the Gen 1B prototype device. h5yjozjewlv5qy71rwa6ojccc78o9ia 80 49 2023-12-13T00:06:53Z Openwaterpete 5 80 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw opw_bloodflow_gen1_hw GitHub repository] for the complete collection of files referenced on this page.''' = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. == Features == * Measure and display blood flow index and blood volume index * Included a hand-held wand, allowing measurement on various scalp areas to identify crucial data points * Data processed on the embedded computer * Movable cart designed for hospital environments = Gen 1 to Gen 1B Device History = The first device sent to a hospital was called a Blood Flow Gen 1 cart, approved by the hospital principal investigator and IRB. Feedback received indicated that the Gen 1 device was challenging to maneuver into small rooms needed for patient measurement. The Gen 1B device was then developed as a reconfigured device that occupied less floor space. Additionally, the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that could mount to an IV pole. The Gen 1B device provided valuable insights into its use within a clinically significant population and was operated by research coordinators or nurse practitioners in a hospital setting. Refer to Gen 1 Blood Flow White Paper for detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is a more detailed and improved version. Gen 1B used large off-the-shelf general components, while Gen 2 employed smaller, less expensive customized components. The data produced from Gen 1B is included in the blood flow Gen 1 analysis repo. One significant hardware result from Gen 1B involved finding design criteria to shift from a wand-based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurements. With the headset, only a few positions on the scalp were observed. [[File:wand gen1b.jpg|thumb|Gen 1B Wand]] For the majority of patients, the best data came from areas on the scalp with no or few hair follicles (e.g., the forehead area). A study was conducted to examine 32 possible positions on the forehead, informing the design for the headset used in Gen 2. Refer to the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] for Gen 2 device designs. Some details of the Gen 1B device are omitted due to the availability of improved versions in the Gen 2 repository. Gen 2 was fully documented and designed as the manufacturable replacement for the Gen 1B prototype device. __TOC__ nm5vhv6xycjj2fsrpjjyjaldbvheda2 81 80 2023-12-13T00:07:14Z Openwaterpete 5 81 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw opw_bloodflow_gen1_hw GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. == Features == * Measure and display blood flow index and blood volume index * Included a hand-held wand, allowing measurement on various scalp areas to identify crucial data points * Data processed on the embedded computer * Movable cart designed for hospital environments = Gen 1 to Gen 1B Device History = The first device sent to a hospital was called a Blood Flow Gen 1 cart, approved by the hospital principal investigator and IRB. Feedback received indicated that the Gen 1 device was challenging to maneuver into small rooms needed for patient measurement. The Gen 1B device was then developed as a reconfigured device that occupied less floor space. Additionally, the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that could mount to an IV pole. The Gen 1B device provided valuable insights into its use within a clinically significant population and was operated by research coordinators or nurse practitioners in a hospital setting. Refer to Gen 1 Blood Flow White Paper for detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is a more detailed and improved version. Gen 1B used large off-the-shelf general components, while Gen 2 employed smaller, less expensive customized components. The data produced from Gen 1B is included in the blood flow Gen 1 analysis repo. One significant hardware result from Gen 1B involved finding design criteria to shift from a wand-based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurements. With the headset, only a few positions on the scalp were observed. [[File:wand gen1b.jpg|thumb|Gen 1B Wand]] For the majority of patients, the best data came from areas on the scalp with no or few hair follicles (e.g., the forehead area). A study was conducted to examine 32 possible positions on the forehead, informing the design for the headset used in Gen 2. Refer to the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] for Gen 2 device designs. Some details of the Gen 1B device are omitted due to the availability of improved versions in the Gen 2 repository. Gen 2 was fully documented and designed as the manufacturable replacement for the Gen 1B prototype device. 1k70kc7ckodihe1ymmtuxdww6znlzdy 90 81 2023-12-13T00:49:37Z Openwaterpete 5 90 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw opw_bloodflow_gen1_hw GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. == Features == * Measure and display blood flow index and blood volume index * Included a hand-held wand, allowing measurement on various scalp areas to identify crucial data points * Data processed on the embedded computer * Movable cart designed for hospital environments = Gen 1 to Gen 1B Device History = <p>The first device sent to a hospital was called a Blood Flow Gen 1 cart, approved by the hospital principal investigator and IRB. Feedback received indicated that the Gen 1 device was challenging to maneuver into small rooms needed for patient measurement. <p/> The Gen 1B device was then developed as a reconfigured device that occupied less floor space. Additionally, the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that could mount to an IV pole. The Gen 1B device provided valuable insights into its use within a clinically significant population and was operated by research coordinators or nurse practitioners in a hospital setting. Refer to Gen 1 Blood Flow White Paper for detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is a more detailed and improved version. Gen 1B used large off-the-shelf general components, while Gen 2 employed smaller, less expensive customized components. The data produced from Gen 1B is included in the blood flow Gen 1 analysis repo. One significant hardware result from Gen 1B involved finding design criteria to shift from a wand-based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurements. With the headset, only a few positions on the scalp were observed. [[File:wand gen1b.jpg|thumb|Gen 1B Wand]] For the majority of patients, the best data came from areas on the scalp with no or few hair follicles (e.g., the forehead area). A study was conducted to examine 32 possible positions on the forehead, informing the design for the headset used in Gen 2. Refer to the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] for Gen 2 device designs. Some details of the Gen 1B device are omitted due to the availability of improved versions in the Gen 2 repository. Gen 2 was fully documented and designed as the manufacturable replacement for the Gen 1B prototype device. c3ti8p6cewwcomromsxt0rhh7jrmjq9 91 90 2023-12-13T00:50:52Z Openwaterpete 5 91 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw opw_bloodflow_gen1_hw GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. == Features == * Measure and display blood flow index and blood volume index * Included a hand-held wand, allowing measurement on various scalp areas to identify crucial data points * Data processed on the embedded computer * Movable cart designed for hospital environments = Gen 1 to Gen 1B Device History = <p>The first device sent to a hospital was called a Blood Flow Gen 1 cart, approved by the hospital principal investigator and IRB. Feedback received indicated that the Gen 1 device was challenging to maneuver into small rooms needed for patient measurement. </p> <p>The Gen 1B device was then developed as a reconfigured device that occupied less floor space. Additionally, the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that could mount to an IV pole. </p> The Gen 1B device provided valuable insights into its use within a clinically significant population and was operated by research coordinators or nurse practitioners in a hospital setting. Refer to Gen 1 Blood Flow White Paper for detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is a more detailed and improved version. Gen 1B used large off-the-shelf general components, while Gen 2 employed smaller, less expensive customized components. The data produced from Gen 1B is included in the blood flow Gen 1 analysis repo. One significant hardware result from Gen 1B involved finding design criteria to shift from a wand-based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurements. With the headset, only a few positions on the scalp were observed. [[File:wand gen1b.jpg|thumb|Gen 1B Wand]] For the majority of patients, the best data came from areas on the scalp with no or few hair follicles (e.g., the forehead area). A study was conducted to examine 32 possible positions on the forehead, informing the design for the headset used in Gen 2. Refer to the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] for Gen 2 device designs. Some details of the Gen 1B device are omitted due to the availability of improved versions in the Gen 2 repository. Gen 2 was fully documented and designed as the manufacturable replacement for the Gen 1B prototype device. 9ikxg5fimkv2ipj2cio5wwoo6g0zlf7 92 91 2023-12-13T00:53:09Z Openwaterpete 5 92 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw opw_bloodflow_gen1_hw GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. == Features == * Measure and display blood flow index and blood volume index * Included a hand-held wand, allowing measurement on various scalp areas to identify crucial data points * Data processed on the embedded computer * Movable cart designed for hospital environments = Gen 1 to Gen 1B Device History = <p>The first device sent to a hospital was called a Blood Flow Gen 1 cart, approved by the hospital principal investigator and IRB. Feedback received indicated that the Gen 1 device was challenging to maneuver into small rooms needed for patient measurement. </p> <p>The Gen 1B device was then developed as a reconfigured device that occupied less floor space. Additionally, the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that could mount to an IV pole. </p> The Gen 1B device provided valuable insights into its use within a clinically significant population and was operated by research coordinators or nurse practitioners in a hospital setting. Refer to Gen 1 Blood Flow White Paper for detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is a more detailed and improved version. Gen 1B used large off-the-shelf general components, while Gen 2 employed smaller, less expensive customized components. The data produced from Gen 1B is included in the blood flow Gen 1 analysis repo. One significant hardware result from Gen 1B involved finding design criteria to shift from a wand-based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurements. With the headset, only a few positions on the scalp were observed. [[File:wand gen1b.jpg|thumb|Gen 1B Wand]] For the majority of patients, the best data came from areas on the scalp with no or few hair follicles (e.g., the forehead area). A study was conducted to examine 32 possible positions on the forehead, informing the design for the headset used in Gen 2. Refer to the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] for Gen 2 device designs. Some details of the Gen 1B device are omitted due to the availability of improved versions in the Gen 2 repository. Gen 2 is fully documented and designed as the manufacturable replacement for the Gen 1B prototype device. oj1rvhuujwzsa5egpsaxnz6nbn42nku 151 92 2023-12-13T13:01:22Z OpenwaterAndrew 3 151 wikitext text/x-wiki Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw opw_bloodflow_gen1_hw GitHub repository] for the complete collection of files referenced on this page. __TOC__ = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. == Features == * Measure and display blood flow index and blood volume index * Included a hand-held wand, allowing measurement on various scalp areas to identify crucial data points * Data processed on the embedded computer * Movable cart designed for hospital environments = Gen 1 to Gen 1B Device History = <p>The first device sent to a hospital was called a Blood Flow Gen 1 cart, approved by the hospital principal investigator and IRB. Feedback received indicated that the Gen 1 device was challenging to maneuver into small rooms needed for patient measurement. </p> <p>The Gen 1B device was then developed as a reconfigured device that occupied less floor space. Additionally, the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that could mount to an IV pole. </p> The Gen 1B device provided valuable insights into its use within a clinically significant population and was operated by research coordinators or nurse practitioners in a hospital setting. Refer to the [https://github.com/OpenwaterInternet/opw_bloodflow_gen1_hw/blob/d66d6d0a634957ca50605b537f23d61bad507b1b/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is a more detailed and improved version. Gen 1B used large off-the-shelf general components, while Gen 2 employed smaller, less expensive customized components. The data produced from Gen 1B is included in the blood flow Gen 1 analysis repo. One significant hardware result from Gen 1B involved finding design criteria to shift from a wand-based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurements. With the headset, only a few positions on the scalp were observed. [[File:wand gen1b.jpg|thumb|Gen 1B Wand]] For the majority of patients, the best data came from areas on the scalp with no or few hair follicles (e.g., the forehead area). A study was conducted to examine 32 possible positions on the forehead, informing the design for the headset used in Gen 2. Refer to the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] for Gen 2 device designs. Some details of the Gen 1B device are omitted due to the availability of improved versions in the Gen 2 repository. Gen 2 is fully documented and designed as the manufacturable replacement for the Gen 1B prototype device. 67axp9fw8jcvfofcbv1udkx4kd5z2uk 152 151 2023-12-13T13:02:00Z OpenwaterAndrew 3 152 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw opw_bloodflow_gen1_hw GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. == Features == * Measure and display blood flow index and blood volume index * Included a hand-held wand, allowing measurement on various scalp areas to identify crucial data points * Data processed on the embedded computer * Movable cart designed for hospital environments = Gen 1 to Gen 1B Device History = <p>The first device sent to a hospital was called a Blood Flow Gen 1 cart, approved by the hospital principal investigator and IRB. Feedback received indicated that the Gen 1 device was challenging to maneuver into small rooms needed for patient measurement. </p> <p>The Gen 1B device was then developed as a reconfigured device that occupied less floor space. Additionally, the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that could mount to an IV pole. </p> The Gen 1B device provided valuable insights into its use within a clinically significant population and was operated by research coordinators or nurse practitioners in a hospital setting. Refer to the [https://github.com/OpenwaterInternet/opw_bloodflow_gen1_hw/blob/d66d6d0a634957ca50605b537f23d61bad507b1b/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is a more detailed and improved version. Gen 1B used large off-the-shelf general components, while Gen 2 employed smaller, less expensive customized components. The data produced from Gen 1B is included in the blood flow Gen 1 analysis repo. One significant hardware result from Gen 1B involved finding design criteria to shift from a wand-based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurements. With the headset, only a few positions on the scalp were observed. [[File:wand gen1b.jpg|thumb|Gen 1B Wand]] For the majority of patients, the best data came from areas on the scalp with no or few hair follicles (e.g., the forehead area). A study was conducted to examine 32 possible positions on the forehead, informing the design for the headset used in Gen 2. Refer to the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] for Gen 2 device designs. Some details of the Gen 1B device are omitted due to the availability of improved versions in the Gen 2 repository. Gen 2 is fully documented and designed as the manufacturable replacement for the Gen 1B prototype device. a28tp7zlldurlj87qk4j11h1iwqa5x3 621 152 2023-12-19T18:38:32Z Admin 1 Protected "[[Blood Flow Gen 1 Hardware]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 152 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen1_hw opw_bloodflow_gen1_hw GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Blood Flow System Gen 1 Hardware Overview = The Generation 1 Stroke Detection Device is the first in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. == Features == * Measure and display blood flow index and blood volume index * Included a hand-held wand, allowing measurement on various scalp areas to identify crucial data points * Data processed on the embedded computer * Movable cart designed for hospital environments = Gen 1 to Gen 1B Device History = <p>The first device sent to a hospital was called a Blood Flow Gen 1 cart, approved by the hospital principal investigator and IRB. Feedback received indicated that the Gen 1 device was challenging to maneuver into small rooms needed for patient measurement. </p> <p>The Gen 1B device was then developed as a reconfigured device that occupied less floor space. Additionally, the laser source module was miniaturized and cost reduced in preparation for a Gen 2 system that could mount to an IV pole. </p> The Gen 1B device provided valuable insights into its use within a clinically significant population and was operated by research coordinators or nurse practitioners in a hospital setting. Refer to the [https://github.com/OpenwaterInternet/opw_bloodflow_gen1_hw/blob/d66d6d0a634957ca50605b537f23d61bad507b1b/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for detailed background and theory of operation. Inside the repository for Gen 1B are visual details of the device in both solid model form and actual photos. The Gen 2 device is a more detailed and improved version. Gen 1B used large off-the-shelf general components, while Gen 2 employed smaller, less expensive customized components. The data produced from Gen 1B is included in the blood flow Gen 1 analysis repo. One significant hardware result from Gen 1B involved finding design criteria to shift from a wand-based design to a static headset design. Holding the sensitive wand manually on the patient's head yielded inconsistent measurements. With the headset, only a few positions on the scalp were observed. [[File:wand gen1b.jpg|thumb|Gen 1B Wand]] For the majority of patients, the best data came from areas on the scalp with no or few hair follicles (e.g., the forehead area). A study was conducted to examine 32 possible positions on the forehead, informing the design for the headset used in Gen 2. Refer to the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] for Gen 2 device designs. Some details of the Gen 1B device are omitted due to the availability of improved versions in the Gen 2 repository. Gen 2 is fully documented and designed as the manufacturable replacement for the Gen 1B prototype device. a28tp7zlldurlj87qk4j11h1iwqa5x3 0 37 131 2023-12-13T03:16:59Z Gvigelet 4 Created page with "== Openwater Gen1 Blood Flow Device Overview ==" 131 wikitext text/x-wiki == Openwater Gen1 Blood Flow Device Overview == 6hl00i9tjcpn928m8q67zo5h9wg6w17 611 131 2023-12-19T04:16:31Z 24.92.36.30 /* Openwater Gen1 Blood Flow Device Overview */ 611 wikitext text/x-wiki == Openwater Gen1 Blood Flow Device Overview == == Overview == The Gen 1 Stroke Detection Device is an advanced medical tool designed for the noninvasive measurement of cerebral blood flow, crucial for stroke detection and neurological research. Building on the foundation of the Gen 1 model, Gen 1B introduces enhanced features and user interface improvements for clinical and research applications. == Principle == The Gen 1 device operates on near-infrared spectroscopy, allowing for sensitive and real-time detection of blood flow changes in the cranial vessels. == Components == # Source Module: Utilizes a precision Thor Labs laser for emitting near-infrared light, modulated for optimal penetration and data acquisition. # Detector Module: Captures light interactions with cranial tissues to quantify blood flow dynamics. # Tower: Houses the primary hardware components, serving as the central hub for data processing and system management. # Wand: A ergonomically designed handheld instrument for precise laser application, incorporating enhanced safety and user feedback features. == Software Architecture == The Gen 1 software is upgraded for improved user experience, featuring: # User-Friendly Interface: Intuitive operation for setting up scans, calibrating, and viewing data. # Enhanced Data Processing: Advanced algorithms for accurate interpretation of cerebral blood flow. # Multi-Mode Operation: Includes modes for alignment, scanning, and backup, with added functionalities for each stage. == Safety Features == Gen 1 emphasizes safety with: * Advanced Laser Safety Board for stringent regulation of laser emissions. * Emergency protocols and interlocks for immediate response to safety concerns. * Compliance with all relevant safety standards. == Operation == Operating the Gen 1 involves: # System Setup: Following clear guidelines for assembling and initializing the device. # Calibration and Alignment: Using the wand for precise targeting and calibration. # Scanning Procedure: Engaging the source module to emit and capture data, with real-time processing for immediate feedback. # Data Management: Efficient access and interpretation of scan data for clinical or research use. == Clinical and Research Application == Gen 1 is designed for both clinical stroke detection and neurological research, offering enhanced capabilities for diverse applications in medical settings. ef4rrgrukos7pd30f9hcqrdkt3l5jr1 622 611 2023-12-19T18:38:45Z Admin 1 Protected "[[Blood Flow Gen 1 Software]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 611 wikitext text/x-wiki == Openwater Gen1 Blood Flow Device Overview == == Overview == The Gen 1 Stroke Detection Device is an advanced medical tool designed for the noninvasive measurement of cerebral blood flow, crucial for stroke detection and neurological research. Building on the foundation of the Gen 1 model, Gen 1B introduces enhanced features and user interface improvements for clinical and research applications. == Principle == The Gen 1 device operates on near-infrared spectroscopy, allowing for sensitive and real-time detection of blood flow changes in the cranial vessels. == Components == # Source Module: Utilizes a precision Thor Labs laser for emitting near-infrared light, modulated for optimal penetration and data acquisition. # Detector Module: Captures light interactions with cranial tissues to quantify blood flow dynamics. # Tower: Houses the primary hardware components, serving as the central hub for data processing and system management. # Wand: A ergonomically designed handheld instrument for precise laser application, incorporating enhanced safety and user feedback features. == Software Architecture == The Gen 1 software is upgraded for improved user experience, featuring: # User-Friendly Interface: Intuitive operation for setting up scans, calibrating, and viewing data. # Enhanced Data Processing: Advanced algorithms for accurate interpretation of cerebral blood flow. # Multi-Mode Operation: Includes modes for alignment, scanning, and backup, with added functionalities for each stage. == Safety Features == Gen 1 emphasizes safety with: * Advanced Laser Safety Board for stringent regulation of laser emissions. * Emergency protocols and interlocks for immediate response to safety concerns. * Compliance with all relevant safety standards. == Operation == Operating the Gen 1 involves: # System Setup: Following clear guidelines for assembling and initializing the device. # Calibration and Alignment: Using the wand for precise targeting and calibration. # Scanning Procedure: Engaging the source module to emit and capture data, with real-time processing for immediate feedback. # Data Management: Efficient access and interpretation of scan data for clinical or research use. == Clinical and Research Application == Gen 1 is designed for both clinical stroke detection and neurological research, offering enhanced capabilities for diverse applications in medical settings. ef4rrgrukos7pd30f9hcqrdkt3l5jr1 0 82 369 2023-12-14T01:36:41Z KedarGrama 6 KedarGrama moved page [[Blood Flow Gen 2 Ananlysis and Classification]] to [[Blood Flow Gen 2 LVO Classification and Analysis]] 369 wikitext text/x-wiki #REDIRECT [[Blood Flow Gen 2 LVO Classification and Analysis]] r3ue8reobqjkwvi4knmc5mlxzqcfsnu 0 2 2 2023-12-12T19:57:09Z OpenwaterAndrew 3 Created page with "Please visit the opw_bloodflow_gen2_hw GitHub repository for the complete collection of files referenced on this page. = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into th..." 2 wikitext text/x-wiki Please visit the opw_bloodflow_gen2_hw GitHub repository for the complete collection of files referenced on this page. = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. Gen 2 Blood Flow System with headset, mounted on IV pole The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. High-level system diagram See the Hardware Architecture Design Document (HADD) and System Architecture Diagram for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. CAD rendering of laser source module assembly See the Source Module folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. See photodiode board folder for and TA mount board folder for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the Assembly and Test for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. 24 V input Outputs 24 V computer output 12 V display output 5 V TEC outputs (2) 24 V and 5 V outputs to tapered amplifier driver 5 V output to seed driver (not used) 12 V output to optical switch (not used) Outputs can be switched on/off via I2C communication Fault monitoring and current/voltage sensing over I2C See PDU folder for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See MLIB folder for electrical design and manufacturing files. See laser control overview for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the Harnessing folder for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the Console Housing folder for details on the mechanical design. CAD rendering of console housing = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power Hybrid cable assembly before installation of cameras and patient contact components One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the Hybrid Cable folder for detailed design and BOM for several cable variants. See the Aggregator Board folder for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the Assembly and Test folder for more detail on assembly and test operations. Fiber Test Assembly contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. Schematic of the Gen2 measurement geometry In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. CAD rendering of Far camera assembly See the Camera Modules folder for more detail on the camera mechanical designs and window specifications. See the Camera Board folder for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the Headset folder. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the Simplified Sensor Module folder (b) Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts. See the Hybrid Cable folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See folder here for accessory design data = Device Operation = Main page: link to software repo See the Device Operation folder for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference Gen 2 Blood Flow Hardware Home System design files Mechanical design files Electrical design files Assembly and test resources Device operation gulvugltajiinktrttjx3915qaq7crz 86 2 2023-12-13T00:36:12Z Openwaterpete 5 86 wikitext text/x-wiki Please visit the https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw repository for the complete collection of files referenced on this page. = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. Gen 2 Blood Flow System with headset, mounted on IV pole The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. High-level system diagram See the Hardware Architecture Design Document (HADD) and System Architecture Diagram for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. CAD rendering of laser source module assembly See the Source Module folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. See photodiode board folder for and TA mount board folder for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the Assembly and Test for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. 24 V input Outputs 24 V computer output 12 V display output 5 V TEC outputs (2) 24 V and 5 V outputs to tapered amplifier driver 5 V output to seed driver (not used) 12 V output to optical switch (not used) Outputs can be switched on/off via I2C communication Fault monitoring and current/voltage sensing over I2C See PDU folder for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See MLIB folder for electrical design and manufacturing files. See laser control overview for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the Harnessing folder for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the Console Housing folder for details on the mechanical design. CAD rendering of console housing = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power Hybrid cable assembly before installation of cameras and patient contact components One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the Hybrid Cable folder for detailed design and BOM for several cable variants. See the Aggregator Board folder for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the Assembly and Test folder for more detail on assembly and test operations. Fiber Test Assembly contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. Schematic of the Gen2 measurement geometry In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. CAD rendering of Far camera assembly See the Camera Modules folder for more detail on the camera mechanical designs and window specifications. See the Camera Board folder for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the Headset folder. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the Simplified Sensor Module folder (b) Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts. See the Hybrid Cable folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See folder here for accessory design data = Device Operation = Main page: link to software repo See the Device Operation folder for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference Gen 2 Blood Flow Hardware Home System design files Mechanical design files Electrical design files Assembly and test resources Device operation nw76a9jhfngen3va6l3z3xebxeljs1c 87 86 2023-12-13T00:37:23Z Openwaterpete 5 87 wikitext text/x-wiki Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw] repository for the complete collection of files referenced on this page. = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. Gen 2 Blood Flow System with headset, mounted on IV pole The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. High-level system diagram See the Hardware Architecture Design Document (HADD) and System Architecture Diagram for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. CAD rendering of laser source module assembly See the Source Module folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. See photodiode board folder for and TA mount board folder for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the Assembly and Test for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. 24 V input Outputs 24 V computer output 12 V display output 5 V TEC outputs (2) 24 V and 5 V outputs to tapered amplifier driver 5 V output to seed driver (not used) 12 V output to optical switch (not used) Outputs can be switched on/off via I2C communication Fault monitoring and current/voltage sensing over I2C See PDU folder for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See MLIB folder for electrical design and manufacturing files. See laser control overview for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the Harnessing folder for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the Console Housing folder for details on the mechanical design. CAD rendering of console housing = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power Hybrid cable assembly before installation of cameras and patient contact components One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the Hybrid Cable folder for detailed design and BOM for several cable variants. See the Aggregator Board folder for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the Assembly and Test folder for more detail on assembly and test operations. Fiber Test Assembly contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. Schematic of the Gen2 measurement geometry In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. CAD rendering of Far camera assembly See the Camera Modules folder for more detail on the camera mechanical designs and window specifications. See the Camera Board folder for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the Headset folder. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the Simplified Sensor Module folder (b) Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts. See the Hybrid Cable folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See folder here for accessory design data = Device Operation = Main page: link to software repo See the Device Operation folder for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference Gen 2 Blood Flow Hardware Home System design files Mechanical design files Electrical design files Assembly and test resources Device operation 06t20nstkox4amflaow84iewxhpj1kt 88 87 2023-12-13T00:38:48Z Openwaterpete 5 88 wikitext text/x-wiki Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page. = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. Gen 2 Blood Flow System with headset, mounted on IV pole The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. High-level system diagram See the Hardware Architecture Design Document (HADD) and System Architecture Diagram for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. CAD rendering of laser source module assembly See the Source Module folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. See photodiode board folder for and TA mount board folder for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the Assembly and Test for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. 24 V input Outputs 24 V computer output 12 V display output 5 V TEC outputs (2) 24 V and 5 V outputs to tapered amplifier driver 5 V output to seed driver (not used) 12 V output to optical switch (not used) Outputs can be switched on/off via I2C communication Fault monitoring and current/voltage sensing over I2C See PDU folder for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See MLIB folder for electrical design and manufacturing files. See laser control overview for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the Harnessing folder for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the Console Housing folder for details on the mechanical design. CAD rendering of console housing = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power Hybrid cable assembly before installation of cameras and patient contact components One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the Hybrid Cable folder for detailed design and BOM for several cable variants. See the Aggregator Board folder for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the Assembly and Test folder for more detail on assembly and test operations. Fiber Test Assembly contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. Schematic of the Gen2 measurement geometry In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. CAD rendering of Far camera assembly See the Camera Modules folder for more detail on the camera mechanical designs and window specifications. See the Camera Board folder for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the Headset folder. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the Simplified Sensor Module folder (b) Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts. See the Hybrid Cable folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See folder here for accessory design data = Device Operation = Main page: link to software repo See the Device Operation folder for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference Gen 2 Blood Flow Hardware Home System design files Mechanical design files Electrical design files Assembly and test resources Device operation 2e35d9jkvlu82ezwyu1277srthtnj83 146 88 2023-12-13T12:36:37Z OpenwaterAndrew 3 146 wikitext text/x-wiki Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page. = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.png|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.png|thumb|High-level system diagram]] See the Hardware Architecture Design Document (HADD) and System Architecture Diagram for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. CAD rendering of laser source module assembly See the Source Module folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. See photodiode board folder for and TA mount board folder for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the Assembly and Test for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. 24 V input Outputs 24 V computer output 12 V display output 5 V TEC outputs (2) 24 V and 5 V outputs to tapered amplifier driver 5 V output to seed driver (not used) 12 V output to optical switch (not used) Outputs can be switched on/off via I2C communication Fault monitoring and current/voltage sensing over I2C See PDU folder for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See MLIB folder for electrical design and manufacturing files. See laser control overview for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the Harnessing folder for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the Console Housing folder for details on the mechanical design. CAD rendering of console housing = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power Hybrid cable assembly before installation of cameras and patient contact components One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the Hybrid Cable folder for detailed design and BOM for several cable variants. See the Aggregator Board folder for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the Assembly and Test folder for more detail on assembly and test operations. Fiber Test Assembly contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. Schematic of the Gen2 measurement geometry In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. CAD rendering of Far camera assembly See the Camera Modules folder for more detail on the camera mechanical designs and window specifications. See the Camera Board folder for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the Headset folder. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the Simplified Sensor Module folder (b) Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts. See the Hybrid Cable folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See folder here for accessory design data = Device Operation = Main page: link to software repo See the Device Operation folder for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference Gen 2 Blood Flow Hardware Home System design files Mechanical design files Electrical design files Assembly and test resources Device operation 8znl14nx1duna33tf4qu0vd735jc4px 148 146 2023-12-13T12:42:57Z OpenwaterAndrew 3 148 wikitext text/x-wiki Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page. = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the Hardware Architecture Design Document (HADD) and System Architecture Diagram for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the Source Module folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. See photodiode board folder for and TA mount board folder for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the Assembly and Test for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. 24 V input Outputs 24 V computer output 12 V display output 5 V TEC outputs (2) 24 V and 5 V outputs to tapered amplifier driver 5 V output to seed driver (not used) 12 V output to optical switch (not used) Outputs can be switched on/off via I2C communication Fault monitoring and current/voltage sensing over I2C See PDU folder for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See MLIB folder for electrical design and manufacturing files. See laser control overview for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the Harnessing folder for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the Console Housing folder for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the Hybrid Cable folder for detailed design and BOM for several cable variants. See the Aggregator Board folder for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the Assembly and Test folder for more detail on assembly and test operations. Fiber Test Assembly contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the Camera Modules folder for more detail on the camera mechanical designs and window specifications. See the Camera Board folder for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the Headset folder. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the Simplified Sensor Module folder [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the Hybrid Cable folder and Assembly and Test folder for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See folder here for accessory design data = Device Operation = Main page: link to software repo See the Device Operation folder for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference Gen 2 Blood Flow Hardware Home System design files Mechanical design files Electrical design files Assembly and test resources Device operation bwkxt91kkot4lih5g5i393um7cel6uu 150 148 2023-12-13T12:59:16Z OpenwaterAndrew 3 150 wikitext text/x-wiki Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page. = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0078%20HADD-002%20Hardware%20Architecture%20Design%20Document.docx%20-%2020231204.pdf Hardware Architecture Design Document (HADD)] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0074-Gen%202.5%20Architecture%20Diagram%201.0-1.pdf System Architecture Diagram] for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/source%20module Source Module folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/photodiode%20board photodiode board folder] for and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/TA%20connector%20board TA mount board folder] for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test] for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. 24 V input Outputs 24 V computer output 12 V display output 5 V TEC outputs (2) 24 V and 5 V outputs to tapered amplifier driver 5 V output to seed driver (not used) 12 V output to optical switch (not used) Outputs can be switched on/off via I2C communication Fault monitoring and current/voltage sensing over I2C See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/PDU PDU folder] for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/MLIB MLIB folder] for electrical design and manufacturing files. See [laser control overview laser control overview] for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/harnessing Harnessing folder] for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/console%20housing Console Housing folder] for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] for detailed design and BOM for several cable variants. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/aggregator%20board Aggregator Board folder] for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on assembly and test operations. [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/fiber%20test%20assembly Fiber Test Assembly] contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/camera%20module Camera Modules folder] for more detail on the camera mechanical designs and window specifications. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/camera%20board Camera Board folder] for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/headset Headset folder]. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/simplified%20sensor%20modules Simplified Sensor Module folder]. [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/accessories folder here] for accessory design data = Device Operation = Main page: link to software repo See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device Operation folder] for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/main Gen 2 Blood Flow Hardware Home] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/system%20design System design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical Mechanical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical Electrical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and test resources] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device operation] 7fmn3auzi5oh5ynffgkc1201w8f2mrv 153 150 2023-12-13T13:02:15Z OpenwaterAndrew 3 153 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page.''' = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0078%20HADD-002%20Hardware%20Architecture%20Design%20Document.docx%20-%2020231204.pdf Hardware Architecture Design Document (HADD)] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0074-Gen%202.5%20Architecture%20Diagram%201.0-1.pdf System Architecture Diagram] for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/source%20module Source Module folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/photodiode%20board photodiode board folder] for and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/TA%20connector%20board TA mount board folder] for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test] for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. 24 V input Outputs 24 V computer output 12 V display output 5 V TEC outputs (2) 24 V and 5 V outputs to tapered amplifier driver 5 V output to seed driver (not used) 12 V output to optical switch (not used) Outputs can be switched on/off via I2C communication Fault monitoring and current/voltage sensing over I2C See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/PDU PDU folder] for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/MLIB MLIB folder] for electrical design and manufacturing files. See [laser control overview laser control overview] for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/harnessing Harnessing folder] for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/console%20housing Console Housing folder] for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] for detailed design and BOM for several cable variants. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/aggregator%20board Aggregator Board folder] for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on assembly and test operations. [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/fiber%20test%20assembly Fiber Test Assembly] contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/camera%20module Camera Modules folder] for more detail on the camera mechanical designs and window specifications. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/camera%20board Camera Board folder] for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/headset Headset folder]. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/simplified%20sensor%20modules Simplified Sensor Module folder]. [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/accessories folder here] for accessory design data = Device Operation = Main page: link to software repo See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device Operation folder] for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/main Gen 2 Blood Flow Hardware Home] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/system%20design System design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical Mechanical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical Electrical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and test resources] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device operation] a78w6p7oilf0ttrtyfnpgbox637ysi2 158 153 2023-12-13T13:45:03Z OpenwaterAndrew 3 /* Blood Flow System Gen 2 Hardware Overview */ 158 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page.''' = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. It is is the 2nd in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. We have used this system to collect a large trove of data from patients in hospital, the summary of this study will be published shortly and a preprint is (link here). We are working on the next generation system now that will be a large shrink and cost reduction to this system. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0078%20HADD-002%20Hardware%20Architecture%20Design%20Document.docx%20-%2020231204.pdf Hardware Architecture Design Document (HADD)] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0074-Gen%202.5%20Architecture%20Diagram%201.0-1.pdf System Architecture Diagram] for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/source%20module Source Module folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/photodiode%20board photodiode board folder] for and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/TA%20connector%20board TA mount board folder] for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test] for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. 24 V input Outputs 24 V computer output 12 V display output 5 V TEC outputs (2) 24 V and 5 V outputs to tapered amplifier driver 5 V output to seed driver (not used) 12 V output to optical switch (not used) Outputs can be switched on/off via I2C communication Fault monitoring and current/voltage sensing over I2C See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/PDU PDU folder] for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/MLIB MLIB folder] for electrical design and manufacturing files. See [laser control overview laser control overview] for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/harnessing Harnessing folder] for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/console%20housing Console Housing folder] for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] for detailed design and BOM for several cable variants. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/aggregator%20board Aggregator Board folder] for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on assembly and test operations. [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/fiber%20test%20assembly Fiber Test Assembly] contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/camera%20module Camera Modules folder] for more detail on the camera mechanical designs and window specifications. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/camera%20board Camera Board folder] for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/headset Headset folder]. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/simplified%20sensor%20modules Simplified Sensor Module folder]. [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/accessories folder here] for accessory design data = Device Operation = Main page: link to software repo See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device Operation folder] for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/main Gen 2 Blood Flow Hardware Home] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/system%20design System design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical Mechanical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical Electrical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and test resources] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device operation] 23p1ce6fndzkn7z4nqy7zexq6hhpnd6 159 158 2023-12-13T13:47:30Z OpenwaterAndrew 3 /* Device Operation */ 159 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page.''' = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. It is is the 2nd in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. We have used this system to collect a large trove of data from patients in hospital, the summary of this study will be published shortly and a preprint is (link here). We are working on the next generation system now that will be a large shrink and cost reduction to this system. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0078%20HADD-002%20Hardware%20Architecture%20Design%20Document.docx%20-%2020231204.pdf Hardware Architecture Design Document (HADD)] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0074-Gen%202.5%20Architecture%20Diagram%201.0-1.pdf System Architecture Diagram] for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/source%20module Source Module folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/photodiode%20board photodiode board folder] for and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/TA%20connector%20board TA mount board folder] for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test] for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. 24 V input Outputs 24 V computer output 12 V display output 5 V TEC outputs (2) 24 V and 5 V outputs to tapered amplifier driver 5 V output to seed driver (not used) 12 V output to optical switch (not used) Outputs can be switched on/off via I2C communication Fault monitoring and current/voltage sensing over I2C See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/PDU PDU folder] for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/MLIB MLIB folder] for electrical design and manufacturing files. See [laser control overview laser control overview] for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/harnessing Harnessing folder] for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/console%20housing Console Housing folder] for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] for detailed design and BOM for several cable variants. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/aggregator%20board Aggregator Board folder] for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on assembly and test operations. [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/fiber%20test%20assembly Fiber Test Assembly] contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/camera%20module Camera Modules folder] for more detail on the camera mechanical designs and window specifications. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/camera%20board Camera Board folder] for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/headset Headset folder]. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/simplified%20sensor%20modules Simplified Sensor Module folder]. [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/accessories folder here] for accessory design data = Device Operation = Main page: [[Blood Flow Gen 2 Software]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device Operation folder] for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/main Gen 2 Blood Flow Hardware Home] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/system%20design System design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical Mechanical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical Electrical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and test resources] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device operation] jbkjw8yegr7x2ddwzvuo2u48dme947l 196 159 2023-12-13T17:11:07Z OpenwaterAndrew 3 /* Power Distribution Unit (PDU) */ 196 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page.''' = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. It is is the 2nd in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. We have used this system to collect a large trove of data from patients in hospital, the summary of this study will be published shortly and a preprint is (link here). We are working on the next generation system now that will be a large shrink and cost reduction to this system. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0078%20HADD-002%20Hardware%20Architecture%20Design%20Document.docx%20-%2020231204.pdf Hardware Architecture Design Document (HADD)] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0074-Gen%202.5%20Architecture%20Diagram%201.0-1.pdf System Architecture Diagram] for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/source%20module Source Module folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/photodiode%20board photodiode board folder] for and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/TA%20connector%20board TA mount board folder] for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test] for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. * 24 V input * Outputs ** 24 V computer output ** 12 V display output ** 5 V TEC outputs (2) ** 24 V and 5 V outputs to tapered amplifier driver ** 5 V output to seed driver (not used) ** 12 V output to optical switch (not used) * Outputs can be switched on/off via I2C communication * Fault monitoring and current/voltage sensing over I2C See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/PDU PDU folder] for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/MLIB MLIB folder] for electrical design and manufacturing files. See [laser control overview laser control overview] for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/harnessing Harnessing folder] for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following - Power on/off switch (Schurter EF12.0035.1110.01) Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/console%20housing Console Housing folder] for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] for detailed design and BOM for several cable variants. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/aggregator%20board Aggregator Board folder] for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on assembly and test operations. [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/fiber%20test%20assembly Fiber Test Assembly] contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/camera%20module Camera Modules folder] for more detail on the camera mechanical designs and window specifications. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/camera%20board Camera Board folder] for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/headset Headset folder]. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/simplified%20sensor%20modules Simplified Sensor Module folder]. [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/accessories folder here] for accessory design data = Device Operation = Main page: [[Blood Flow Gen 2 Software]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device Operation folder] for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/main Gen 2 Blood Flow Hardware Home] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/system%20design System design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical Mechanical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical Electrical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and test resources] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device operation] 9d6lkcb4ad3qnu5ku22w8ftp7b78mjs 197 196 2023-12-13T17:11:32Z OpenwaterAndrew 3 /* Buttons and Switches */ 197 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page.''' = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. It is is the 2nd in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. We have used this system to collect a large trove of data from patients in hospital, the summary of this study will be published shortly and a preprint is (link here). We are working on the next generation system now that will be a large shrink and cost reduction to this system. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0078%20HADD-002%20Hardware%20Architecture%20Design%20Document.docx%20-%2020231204.pdf Hardware Architecture Design Document (HADD)] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0074-Gen%202.5%20Architecture%20Diagram%201.0-1.pdf System Architecture Diagram] for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/source%20module Source Module folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/photodiode%20board photodiode board folder] for and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/TA%20connector%20board TA mount board folder] for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test] for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. * 24 V input * Outputs ** 24 V computer output ** 12 V display output ** 5 V TEC outputs (2) ** 24 V and 5 V outputs to tapered amplifier driver ** 5 V output to seed driver (not used) ** 12 V output to optical switch (not used) * Outputs can be switched on/off via I2C communication * Fault monitoring and current/voltage sensing over I2C See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/PDU PDU folder] for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/MLIB MLIB folder] for electrical design and manufacturing files. See [laser control overview laser control overview] for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/harnessing Harnessing folder] for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following: * Power on/off switch (Schurter EF12.0035.1110.01) * Emergency stop switch (Omron A22E-S-11) === Data Ports === USB-C data download port for connecting flash drives on left side of device USB-B system debug port on bottom face Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/console%20housing Console Housing folder] for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] for detailed design and BOM for several cable variants. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/aggregator%20board Aggregator Board folder] for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on assembly and test operations. [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/fiber%20test%20assembly Fiber Test Assembly] contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/camera%20module Camera Modules folder] for more detail on the camera mechanical designs and window specifications. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/camera%20board Camera Board folder] for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/headset Headset folder]. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/simplified%20sensor%20modules Simplified Sensor Module folder]. [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/accessories folder here] for accessory design data = Device Operation = Main page: [[Blood Flow Gen 2 Software]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device Operation folder] for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/main Gen 2 Blood Flow Hardware Home] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/system%20design System design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical Mechanical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical Electrical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and test resources] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device operation] 9bxrjujkrup935ss1burvs91hjdrxf4 198 197 2023-12-13T17:11:46Z OpenwaterAndrew 3 /* Data Ports */ 198 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page.''' = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. It is is the 2nd in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. We have used this system to collect a large trove of data from patients in hospital, the summary of this study will be published shortly and a preprint is (link here). We are working on the next generation system now that will be a large shrink and cost reduction to this system. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0078%20HADD-002%20Hardware%20Architecture%20Design%20Document.docx%20-%2020231204.pdf Hardware Architecture Design Document (HADD)] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0074-Gen%202.5%20Architecture%20Diagram%201.0-1.pdf System Architecture Diagram] for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/source%20module Source Module folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/photodiode%20board photodiode board folder] for and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/TA%20connector%20board TA mount board folder] for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test] for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. * 24 V input * Outputs ** 24 V computer output ** 12 V display output ** 5 V TEC outputs (2) ** 24 V and 5 V outputs to tapered amplifier driver ** 5 V output to seed driver (not used) ** 12 V output to optical switch (not used) * Outputs can be switched on/off via I2C communication * Fault monitoring and current/voltage sensing over I2C See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/PDU PDU folder] for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/MLIB MLIB folder] for electrical design and manufacturing files. See [laser control overview laser control overview] for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/harnessing Harnessing folder] for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following: * Power on/off switch (Schurter EF12.0035.1110.01) * Emergency stop switch (Omron A22E-S-11) === Data Ports === * USB-C data download port for connecting flash drives on left side of device * USB-B system debug port on bottom face * Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/console%20housing Console Housing folder] for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] for detailed design and BOM for several cable variants. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/aggregator%20board Aggregator Board folder] for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on assembly and test operations. [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/fiber%20test%20assembly Fiber Test Assembly] contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/camera%20module Camera Modules folder] for more detail on the camera mechanical designs and window specifications. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/camera%20board Camera Board folder] for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/headset Headset folder]. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/simplified%20sensor%20modules Simplified Sensor Module folder]. [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/accessories folder here] for accessory design data = Device Operation = Main page: [[Blood Flow Gen 2 Software]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device Operation folder] for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/main Gen 2 Blood Flow Hardware Home] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/system%20design System design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical Mechanical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical Electrical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and test resources] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device operation] 91ayt37pogcz5bbsjcba1c0c52v0g25 555 198 2023-12-16T00:03:15Z KedarGrama 6 /* Blood Flow System Gen 2 Hardware Overview */ 555 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page.''' = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. It is is the 2nd in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. We have used this system to collect a large trove of data from patients in hospital, the summary of this study will be published shortly and a preprint is [https://medrxiv.org/cgi/content/short/2023.12.14.23299992v1 available here]. We are working on the next generation system now that will be a large shrink and cost reduction to this system. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0078%20HADD-002%20Hardware%20Architecture%20Design%20Document.docx%20-%2020231204.pdf Hardware Architecture Design Document (HADD)] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0074-Gen%202.5%20Architecture%20Diagram%201.0-1.pdf System Architecture Diagram] for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/source%20module Source Module folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/photodiode%20board photodiode board folder] for and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/TA%20connector%20board TA mount board folder] for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test] for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. * 24 V input * Outputs ** 24 V computer output ** 12 V display output ** 5 V TEC outputs (2) ** 24 V and 5 V outputs to tapered amplifier driver ** 5 V output to seed driver (not used) ** 12 V output to optical switch (not used) * Outputs can be switched on/off via I2C communication * Fault monitoring and current/voltage sensing over I2C See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/PDU PDU folder] for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/MLIB MLIB folder] for electrical design and manufacturing files. See [laser control overview laser control overview] for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/harnessing Harnessing folder] for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following: * Power on/off switch (Schurter EF12.0035.1110.01) * Emergency stop switch (Omron A22E-S-11) === Data Ports === * USB-C data download port for connecting flash drives on left side of device * USB-B system debug port on bottom face * Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/console%20housing Console Housing folder] for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] for detailed design and BOM for several cable variants. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/aggregator%20board Aggregator Board folder] for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on assembly and test operations. [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/fiber%20test%20assembly Fiber Test Assembly] contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/camera%20module Camera Modules folder] for more detail on the camera mechanical designs and window specifications. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/camera%20board Camera Board folder] for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/headset Headset folder]. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/simplified%20sensor%20modules Simplified Sensor Module folder]. [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/accessories folder here] for accessory design data = Device Operation = Main page: [[Blood Flow Gen 2 Software]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device Operation folder] for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/main Gen 2 Blood Flow Hardware Home] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/system%20design System design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical Mechanical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical Electrical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and test resources] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device operation] e67kw6sv6zqgj6lja7ioj56bl9h3ua9 624 555 2023-12-19T18:39:43Z Admin 1 Protected "[[Blood Flow Gen 2 Hardware]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 555 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page.''' = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. It is is the 2nd in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. We have used this system to collect a large trove of data from patients in hospital, the summary of this study will be published shortly and a preprint is [https://medrxiv.org/cgi/content/short/2023.12.14.23299992v1 available here]. We are working on the next generation system now that will be a large shrink and cost reduction to this system. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0078%20HADD-002%20Hardware%20Architecture%20Design%20Document.docx%20-%2020231204.pdf Hardware Architecture Design Document (HADD)] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0074-Gen%202.5%20Architecture%20Diagram%201.0-1.pdf System Architecture Diagram] for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/source%20module Source Module folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/photodiode%20board photodiode board folder] for and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/TA%20connector%20board TA mount board folder] for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test] for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. * 24 V input * Outputs ** 24 V computer output ** 12 V display output ** 5 V TEC outputs (2) ** 24 V and 5 V outputs to tapered amplifier driver ** 5 V output to seed driver (not used) ** 12 V output to optical switch (not used) * Outputs can be switched on/off via I2C communication * Fault monitoring and current/voltage sensing over I2C See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/PDU PDU folder] for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/MLIB MLIB folder] for electrical design and manufacturing files. See [laser control overview laser control overview] for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/harnessing Harnessing folder] for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following: * Power on/off switch (Schurter EF12.0035.1110.01) * Emergency stop switch (Omron A22E-S-11) === Data Ports === * USB-C data download port for connecting flash drives on left side of device * USB-B system debug port on bottom face * Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/console%20housing Console Housing folder] for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] for detailed design and BOM for several cable variants. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/aggregator%20board Aggregator Board folder] for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on assembly and test operations. [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/fiber%20test%20assembly Fiber Test Assembly] contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/camera%20module Camera Modules folder] for more detail on the camera mechanical designs and window specifications. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/camera%20board Camera Board folder] for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/headset Headset folder]. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/simplified%20sensor%20modules Simplified Sensor Module folder]. [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/accessories folder here] for accessory design data = Device Operation = Main page: [[Blood Flow Gen 2 Software]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device Operation folder] for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} Hardware Resource Quick Reference [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/main Gen 2 Blood Flow Hardware Home] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/system%20design System design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical Mechanical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical Electrical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and test resources] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device operation] e67kw6sv6zqgj6lja7ioj56bl9h3ua9 664 624 2023-12-22T19:45:00Z KedarGrama 6 /* High-Level Specifications and BOM */ 664 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw opw_bloodflow_gen2_hw GitHub repository] repository for the complete collection of files referenced on this page.''' = Blood Flow System Gen 2 Hardware Overview = The Openwater Generation 2 Blood Flow system is a blood flow measurement device. It is is the 2nd in a series of prototypes developed to measure blood flow in the brain of a human. Data from Generation 1 Stroke Detection Device testing in human studies informed design changes to subsequent prototypes. We have used this system to collect a large trove of data from patients in hospital, the summary of this study will be published shortly and a preprint is [https://medrxiv.org/cgi/content/short/2023.12.14.23299992v1 available here]. We are working on the next generation system now that will be a large shrink and cost reduction to this system. Openwater’s blood flow technology uses near-infrared laser light, combined with camera sensors, to measure blood flow. Short pulses of laser light are transmitted to the patient through optical fibers and are injected into the tissue. After diffusing through the tissue some of the remitted light is collected by camera sensors. Images collected by the cameras are analyzed, and information about blood flow volume and rate are extracted. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/gen1%20BF%20White%20Paper.pdf Gen 1 Blood Flow White Paper] for more detailed background and theory of operation. Openwater has developed multiple generations of blood flow devices to perform this measurement. The Generation 1 system required a large cart assembly for electronics, and used a handheld wand to interface with the patient. In Generation 2, the system has been miniaturized into a package which can be easily moved through a clinical or operating room, and can be mounted to an IV pole. Instead of a wand, the laser optics and camera sensors are fixed to the patient with a strap to improve the stability of the measurement. [[File:System.PNG|thumb|Gen 2 Blood Flow System with headset, mounted on IV pole]] The Gen 2 Blood Flow system is composed of a console assembly and patient contact cable assembly. The console contains the system electronics and houses a touchscreen display for operation, buttons for power on/off and emergency stop, as well as several ports for downloading data and performing debug operations. The patient contact cable assembly is a hybrid cable containing both electrical connections and optical fiber, terminated with sensor module assemblies which attach to the patient. The laser and camera optics which interface with the patient can be configured as needed for the specific application. For example, Openwater has developed full headsets for collecting measurements on the forehead, as well as very simple modules which can be placed on the forearm or torso. See the Patient Contact Cable Assembly section below for additional detail. [[File:System_diagram.PNG|thumb|High-level system diagram]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0078%20HADD-002%20Hardware%20Architecture%20Design%20Document.docx%20-%2020231204.pdf Hardware Architecture Design Document (HADD)] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/blob/26394e010a58ca1bcea91e197601f0917618e86e/system%20design/D0074-Gen%202.5%20Architecture%20Diagram%201.0-1.pdf System Architecture Diagram] for more information about the configuration of major hardware components. = Console Assembly = == Laser Source Assembly == Laser light of the appropriate wavelength, power, and pulse characteristics is central to the operation of the blood flow system. The Laser Source Assembly produces these laser pulses, and couples the light into optical fiber for transmission to the patient. The laser subsystem is based on a distributed feedback (DFB) laser which feeds into a tapered amplifier (TA) for high power pulsing. After being emitted from the TA as a free space beam, the laser light is coupled into an optical fiber, and split into two output channels. The various components are packaged on an aluminum baseplate, which is mounted within the system console. The laser components and onboard thermoelectric coolers are driven by off-the-shelf laser drivers. Control of the lasers and coolers is performed by the Multipurpose Laser Interface Board, which is covered in additional detail below. In addition to the components required to generate laser pulses, the Laser Source Assembly includes a photodiode circuit for continuously monitoring the emitted laser light. Signals from this photodiode are sent through several safety circuits within the Multipurpose Laser Interface Board, which shuts off the laser system in the event of aberrant behavior. [[File:Source module.PNG|thumb|CAD rendering of laser source module assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/source%20module Source Module folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/photodiode%20board photodiode board folder] for and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/TA%20connector%20board TA mount board folder] for detail on custom electrical board assemblies contained within the Source Module. == Computer == The Gen 2 Blood Flow System uses a computer module from D3 Engineering, based on a TDA4 processor unit. The computer is outfitted with an expansion board that allows up to 8 or 16 cameras to be connected at once via individual FPDLink III connectors. Documentation for this device is available from the supplier. == Electronics Bracket Assembly == The Electronics Bracket Assembly includes the majority of custom electrical components within the console. These include Power Distribution Unit (PDU) Multipurpose Laser Interface Board (MLIB) Laser drivers AC-DC converter Mechanically, these electrical subassemblies are assembled on a sheet metal bracket, which in turn is mounted on the console baseplate. See console assembly instructions in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test] for details on its construction. A more detailed description of its components is included below. === Power Distribution Unit (PDU) === The PDU conditions and distributes power to various electrical components within the console assembly. * 24 V input * Outputs ** 24 V computer output ** 12 V display output ** 5 V TEC outputs (2) ** 24 V and 5 V outputs to tapered amplifier driver ** 5 V output to seed driver (not used) ** 12 V output to optical switch (not used) * Outputs can be switched on/off via I2C communication * Fault monitoring and current/voltage sensing over I2C See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/PDU PDU folder] for electrical design and manufacturing files. === Multipurpose Laser Interface Board (MLIB) === The MLIB controls and ensures the safe operation of the laser system. This board has connections to both the seed laser and tapered amplifier (TA) for both control and monitoring functions. Thermoelectric coolers on the seed and TA components are also controlled by the MLIB. Ensuring safe operation of the laser source assembly is a primary function of the MLIB. Redundant safety circuits monitor electrical signals from the tapered amplifier, as well as optical feedback signals generated by the source module photodiode circuit. Laser power is shut off in the event of aberrant laser behavior or errors in the laser driving signals. In addition to safety monitoring by the MLIB, device operators can depress an emergency stop button on the console, which shuts off power to the laser source. See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/MLIB MLIB folder] for electrical design and manufacturing files. See [laser control overview laser control overview] for high level operation === Laser Drivers === Commercially available laser drivers are used to drive the Seed and Tapered Amplifier laser components within the source module. Wavelength Electronics MPL500 for seed Wavelength Electronics LD10CHA for tapered amplifier Documentation for these devices is available from the supplier. AC-DC Converter A commercially available AC-DC converter (Cosel GHA500F) is used to convert from standard 120 V outlet voltage to 24 V for supply to the Power Distribution Unit. Documentation for this device is available from the supplier. Additional Electrical Subassemblies and Interfaces === Display === A touch screen display is the primary interface for operating the blood flow system (Future Designs ELI101-IPHW). Documentation for this device is available from the supplier. === Harnessing === System harnessing is mostly custom, with the exception of several commercially available cables for basic power and communication functions. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/harnessing Harnessing folder] for the system harnessing diagram and detailed harness drawings, as well as a list of commercially available cables used within the system. === Buttons and Switches === The console assembly features the following: * Power on/off switch (Schurter EF12.0035.1110.01) * Emergency stop switch (Omron A22E-S-11) === Data Ports === * USB-C data download port for connecting flash drives on left side of device * USB-B system debug port on bottom face * Ethernet system debug port on bottom face == Console Housing == Console housing features: Handle for easy carrying Standard VESA mounting hole pattern for direct mounting to display mounting brackets, or to accessory brackets for attachment to IV pole See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/console%20housing Console Housing folder] for details on the mechanical design. [[File:Console housing.PNG|thumb|CAD rendering of console housing]] = Patient Contact Cable Assembly = == Hybrid Cable Assembly == The Patient Contact Cable Assembly includes optical and electrical components required to transmit laser light to the patient, and to detect light for analysis after it has interacted with the tissue. The cable itself includes the following components: optical fibers and diffusing optics camera assemblies, including camera boards and window optics aggregator board assembly for conditioning camera power [[File:Hybrid cable.PNG|thumb|Hybrid cable assembly before installation of cameras and patient contact components]] One critical function of the hybrid cable is delivering diffuse laser light to the patient. Laser fiber optics are packaged behind a diffusing element, which broadly spreads the light as it is emitted from the end of the cable. The optical assembly and beam characteristics have been carefully designed and tested to ensure that the beam does not create a damage risk to skin or eyes. The Gen 2 blood flow headset is a Class 1 laser device, which can be operated without personal protective equipment or additional procedural safety controls. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] for detailed design and BOM for several cable variants. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/aggregator%20board Aggregator Board folder] for detail on the electrical design and manufacturing of the board assembly which resides within the cable. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on assembly and test operations. [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/fiber%20test%20assembly Fiber Test Assembly] contains mechanical detail of the test assembly used during manufacturing. == Cameras == Blood Flow System cameras are built using a Himax HM5530 rolling shutter CMOS image sensor. The camera module is composed of a rigid-flex board assembly (to which the image sensor is die attached and wire bonded), an injection molded plastic housing, and an optical window. During measurement, the optical window makes contact with the skin surface. The window has a bandpass coating designed to transmit only the laser wavelength and minimize the effect of ambient light on the measurement. A well-defined aperture at the window surface, together with a well defined window-sensor distance, defines the expected size of the laser speckle pattern at the image sensor focal plane. Two variants are used in the Gen 2 system, referred to as Near and Far cameras. The housings and window apertures are designed to create similar laser speckle sizes at two different aperture to detector distances. Far cameras have a large aperture, and corresponding short window to sensor distance. The laser light has been absorbed and scattered through a relatively large volume of tissue so the far camera is designed to gather as much light as possible. [[File:Gen2 meas geo.PNG|thumb|Schematic of the Gen2 measurement geometry]] In order to make the amount of light detected on the near cameras similar to that of the Far cameras, their apertures are smaller. This avoids saturating the near camera sensors, since the amount of scattered light is higher close to the laser source. The window to sensor distance is longer for the Near cameras to further reduce the amount of light detected. As a result, the speckle size on the near cameras is larger than on the far cameras. [[File:Camera.PNG|thumb|CAD rendering of Far camera assembly]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/camera%20module Camera Modules folder] for more detail on the camera mechanical designs and window specifications. See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical/camera%20board Camera Board folder] for detail on electrical design and manufacturing of the rigid-flex board. == Patient Contact Sensor Modules == The fiber tips and cameras themselves can be packaged in a variety of ways to interface with regions of interest on the body. Two example designs: 6-camera headset for symmetrical positioning on the forehead. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/headset Headset folder]. Simplified individual module design, 2 cameras each, for positioning on various body parts. See detailed design files in the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/simplified%20sensor%20modules Simplified Sensor Module folder]. [[File:Headset and module.PNG|thumb|Examples of sensor configurations. a) Full headset with 6 cameras for simultaneous and symmetrical measurements at the forehead. b) Simplified 2-camera module for positioning on a wider range of body parts.]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/hybrid%20cable Hybrid Cable folder] and [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and Test folder] for more detail on design, assembly, and test of this subsystem. = Accessories = Several accessories can be utilized when collecting measurements with the blood flow system IV pole - GCX FLP-0001-61E series Veils for minimizing ambient light Straps of various lengths for using simplified sensor modules on various body parts See [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical/major%20subassemblies/accessories folder here] for accessory design data = Device Operation = Main page: [[Blood Flow Gen 2 Software]] See the [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device Operation folder] for detail on operating the Blood Flow Device: = High-Level Specifications and BOM = {| class="wikitable" |- | Laser wavelength || 785 nm |- | Laser source configuration|| DFB seed + tapered amplifier |- | Number of camera channels|| 6 is standard and up to 16 |- | Laser fiber channels || 2 |- | Console interface || Touch screen |- | Console mounting hole pattern || VESA 75 |- | Device weight || Approx. 14kg |- | Console dimensions || 50 x 23 x 35 cm (W x D x H) |- | Laser seed || Eagleyard EYP-DFB-0785-00012-1500-BFY22-0007 |- | Laser tapered amplifier || Eagleyard EYP-TPA-0780-03000-4006-BFU09-0011 |- | Image sensors || Himax HM5530 CMOS |- | Free space isolator || EOT ISO-FRDY-04-785-N ISOLATOR |- | Computer || D3 TDA4 stack 16 channel fpd-link III |- | Laser seed driver || Wavelength Electronics MPL500 |- | Laser TA driver || Wavelength Electronics LD10CHA |- | Power entry module || Schurter EF12.0035.1110.01 |- | AC-DC converter || Cosel GHA500F |- | Display || FDI ELI101-IPHW |- | Thermal controllers || Wavelength Electronics WTC3293/WTC3243/WEV300 |} ==== Hardware Resource Quick Reference ==== [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/main Gen 2 Blood Flow Hardware Home] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/system%20design System design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/mechanical Mechanical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/electrical Electrical design files] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/assembly%20and%20test Assembly and test resources] [https://github.com/OpenwaterInternet/opw_bloodflow_gen2_hw/tree/26394e010a58ca1bcea91e197601f0917618e86e/device%20operation Device operation] ==== Software Quick Reference ==== Software [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw repository], [[Blood Flow Gen 1 Software|wiki]] Machine learning [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai repository], [[Blood Flow Gen 2 Ananlysis and Classification|wiki]] qlg6grsp1x2bevu9zpn457dgkix9laf 0 67 281 2023-12-13T21:43:40Z KedarGrama 6 Created page with "For classification, we use the deep learning model in [1]. The model is based on a transformer architecture that learns discriminative feature representations from the raw blood flow waveform. Transformers are a type of neural network that are well-suited for natural language processing tasks, but they can also be used for other tasks, such as blood flow classification. To address the problem of data insufficiency, we pre-trained the network to reconstruct the wavefor..." 281 wikitext text/x-wiki For classification, we use the deep learning model in [1]. The model is based on a transformer architecture that learns discriminative feature representations from the raw blood flow waveform. Transformers are a type of neural network that are well-suited for natural language processing tasks, but they can also be used for other tasks, such as blood flow classification. To address the problem of data insufficiency, we pre-trained the network to reconstruct the waveforms from 80 scans on healthy volunteers. We added a set of 4 transformer decoders to the network for reconstruction. After pre-training, we discarded the decoder stack of the network. amgyegl87uuxm6qufh93kphn5eutfk2 368 281 2023-12-14T01:36:41Z KedarGrama 6 KedarGrama moved page [[Blood Flow Gen 2 Ananlysis and Classification]] to [[Blood Flow Gen 2 LVO Classification and Analysis]] 281 wikitext text/x-wiki For classification, we use the deep learning model in [1]. The model is based on a transformer architecture that learns discriminative feature representations from the raw blood flow waveform. Transformers are a type of neural network that are well-suited for natural language processing tasks, but they can also be used for other tasks, such as blood flow classification. To address the problem of data insufficiency, we pre-trained the network to reconstruct the waveforms from 80 scans on healthy volunteers. We added a set of 4 transformer decoders to the network for reconstruction. After pre-training, we discarded the decoder stack of the network. amgyegl87uuxm6qufh93kphn5eutfk2 375 368 2023-12-14T18:35:42Z 135.180.195.174 save work 375 wikitext text/x-wiki Our Laser speckle contrast imaging (LSCI) device gives us relative blood flow measurements. We measure the blood flow at three positions on the forehead - near(short spacing to superficial flow), horizontal(on the forehead) and vertical(on the lower temple) which roughly interrogate flows in the volumes supplied by the ICA and MCA. The data acquisition on the patient acquires 15 seconds of flow of data simultaneously on the left and right at each position. For a majority of the subjects in the LVO study (reminder:link) the order of acquisition was in the order shown in the plots below. We experimented with traditional methods of feature extraction and classification based on published literature in the domain. We also implemented deep learning methods which are described below. === Feature extraction and classification === We implement a large set of features in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures] class in our code. The features we implement are based on published literature [1],[2],[3]. The features are pulse area under the curve, pulse area under the curve to P1[1], amplitude, average, modulation depth(equivalent to pulsatility index)[3], skewness, kurtosis, pulse canopy[1], pulse onset(time to P1)[2] and Velocity Curvature Index[2]. We experimented with Velocity Asymmetry Index[4] but it did not help with the classification accuracy. We implemented these features to be computed on the average pulse in a scan shown in the figure below. We also implemented it to be computed on every pulse in a scan and used the median and the range in a scan as classification feature and this is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The classification yielded a sensitivity of ~68% at a specificity of 80% with an AUC around 0.72 which is comparable to measures used in the field today like RACE and LAMS. [1] Thorpe, S. G., Thibeault, C. M., Canac, N., Jalaleddini, K., Dorn, A., Wilk, S. J., ... & Hamilton, R. B. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', ''15''(2), e0228642. [2] Thorpe, S. G., Thibeault, C. M., Wilk, S. J., O’Brien, M., Canac, N., Ranjbaran, M., ... & Hamilton, R. B. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', ''10'', 475-484. [3] Jalaleddini, K., Canac, N., Thorpe, S. G., O’Brien, M. J., Ranjbaran, M., Delay, B., ... & Hamilton, R. B. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', ''67''(3), 883-892. [4] Thorpe, S. G., Thibeault, C. M., Canac, N., Wilk, S. J., Devlin, T., & Hamilton, R. B. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', ''9'', 847. 6kj56hn41rqp1tdzedmb911saaldu5n 376 375 2023-12-14T18:52:50Z 135.180.195.174 376 wikitext text/x-wiki ## Laser Speckle Contrast Imaging (LSCI) for Large Vessel Occlusion (LVO) Detection: A MediaWiki Edit **Introduction** {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} This article describes the use of Openwater's Gen2 blood flow device which uses Laser Speckle Contrast Imaging (LSCI) to measure relative blood flow in the context of detecting Large Vessel Occlusion (LVO). It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. **Data Acquisition** * The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) * These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). * Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. * The acquisition order follows a specific pattern for most LVO study participants (see [[link to LVO study]]). **Feature Extraction and Classification** * A large set of features is implemented in the PulseFeatures class, based on published literature ([[1]], [[2]], [[3]]). * These features include: * Pulse area under the curve (AUC) * Pulse AUC to P1 peak * Amplitude * Average * Modulation depth (pulsatility index) * Skewness * Kurtosis * Pulse canopy * Pulse onset time * Velocity Curvature Index * Feature calculation is performed on: * The average pulse in a scan * Each individual pulse in a scan (median and range used for classification) * This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures] * The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of 0.72. * This performance is comparable to existing LVO detection methods like RACE and LAMS. **References** * [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. *PloS one*, 15(2), e0228642. * [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. *Translational stroke research*, 10, 475-484. * [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. *IEEE Transactions on Biomedical Engineering*, 67(3), 883-892. **Note:** This is a MediaWiki-formatted version of the original text, ready for inclusion in a wiki page. You can customize the formatting and add additional content as needed. Our Laser speckle contrast imaging (LSCI) device gives us relative blood flow measurements. We measure the blood flow at three positions on the forehead - near(short spacing to superficial flow), horizontal(on the forehead) and vertical(on the lower temple) which roughly interrogate flows in the volumes supplied by the ICA and MCA. The data acquisition on the patient acquires 15 seconds of flow of data simultaneously on the left and right at each position. For a majority of the subjects in the LVO study (reminder:link) the order of acquisition was in the order shown in the plots below. We experimented with traditional methods of feature extraction and classification based on published literature in the domain. We also implemented deep learning methods which are described below. === Feature extraction and classification === We implement a large set of features in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures] class in our code. The features we implement are based on published literature [1],[2],[3]. The features are pulse area under the curve, pulse area under the curve to P1[1], amplitude, average, modulation depth(equivalent to pulsatility index)[3], skewness, kurtosis, pulse canopy[1], pulse onset(time to P1)[2] and Velocity Curvature Index[2]. We experimented with Velocity Asymmetry Index[4] but it did not help with the classification accuracy. We implemented these features to be computed on the average pulse in a scan shown in the figure below. We also implemented it to be computed on every pulse in a scan and used the median and the range in a scan as classification feature and this is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The classification yielded a sensitivity of ~68% at a specificity of 80% with an AUC around 0.72 which is comparable to measures used in the field today like RACE and LAMS. [1] Thorpe, S. G., Thibeault, C. M., Canac, N., Jalaleddini, K., Dorn, A., Wilk, S. J., ... & Hamilton, R. B. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', ''15''(2), e0228642. [2] Thorpe, S. G., Thibeault, C. M., Wilk, S. J., O’Brien, M., Canac, N., Ranjbaran, M., ... & Hamilton, R. B. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', ''10'', 475-484. [3] Jalaleddini, K., Canac, N., Thorpe, S. G., O’Brien, M. J., Ranjbaran, M., Delay, B., ... & Hamilton, R. B. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', ''67''(3), 883-892. [4] Thorpe, S. G., Thibeault, C. M., Canac, N., Wilk, S. J., Devlin, T., & Hamilton, R. B. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', ''9'', 847. 33gi6be5umxf7gkxlmfzxmh02l6cnl0 377 376 2023-12-14T18:53:13Z 135.180.195.174 377 wikitext text/x-wiki **Introduction** {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} This article describes the use of Openwater's Gen2 blood flow device which uses Laser Speckle Contrast Imaging (LSCI) to measure relative blood flow in the context of detecting Large Vessel Occlusion (LVO). It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. **Data Acquisition** * The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) * These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). * Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. * The acquisition order follows a specific pattern for most LVO study participants (see [[link to LVO study]]). **Feature Extraction and Classification** * A large set of features is implemented in the PulseFeatures class, based on published literature ([[1]], [[2]], [[3]]). * These features include: * Pulse area under the curve (AUC) * Pulse AUC to P1 peak * Amplitude * Average * Modulation depth (pulsatility index) * Skewness * Kurtosis * Pulse canopy * Pulse onset time * Velocity Curvature Index * Feature calculation is performed on: * The average pulse in a scan * Each individual pulse in a scan (median and range used for classification) * This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures] * The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of 0.72. * This performance is comparable to existing LVO detection methods like RACE and LAMS. **References** * [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. *PloS one*, 15(2), e0228642. * [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. *Translational stroke research*, 10, 475-484. * [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. *IEEE Transactions on Biomedical Engineering*, 67(3), 883-892. **Note:** This is a MediaWiki-formatted version of the original text, ready for inclusion in a wiki page. You can customize the formatting and add additional content as needed. Our Laser speckle contrast imaging (LSCI) device gives us relative blood flow measurements. We measure the blood flow at three positions on the forehead - near(short spacing to superficial flow), horizontal(on the forehead) and vertical(on the lower temple) which roughly interrogate flows in the volumes supplied by the ICA and MCA. The data acquisition on the patient acquires 15 seconds of flow of data simultaneously on the left and right at each position. For a majority of the subjects in the LVO study (reminder:link) the order of acquisition was in the order shown in the plots below. We experimented with traditional methods of feature extraction and classification based on published literature in the domain. We also implemented deep learning methods which are described below. === Feature extraction and classification === We implement a large set of features in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures] class in our code. The features we implement are based on published literature [1],[2],[3]. The features are pulse area under the curve, pulse area under the curve to P1[1], amplitude, average, modulation depth(equivalent to pulsatility index)[3], skewness, kurtosis, pulse canopy[1], pulse onset(time to P1)[2] and Velocity Curvature Index[2]. We experimented with Velocity Asymmetry Index[4] but it did not help with the classification accuracy. We implemented these features to be computed on the average pulse in a scan shown in the figure below. We also implemented it to be computed on every pulse in a scan and used the median and the range in a scan as classification feature and this is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The classification yielded a sensitivity of ~68% at a specificity of 80% with an AUC around 0.72 which is comparable to measures used in the field today like RACE and LAMS. [1] Thorpe, S. G., Thibeault, C. M., Canac, N., Jalaleddini, K., Dorn, A., Wilk, S. J., ... & Hamilton, R. B. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', ''15''(2), e0228642. [2] Thorpe, S. G., Thibeault, C. M., Wilk, S. J., O’Brien, M., Canac, N., Ranjbaran, M., ... & Hamilton, R. B. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', ''10'', 475-484. [3] Jalaleddini, K., Canac, N., Thorpe, S. G., O’Brien, M. J., Ranjbaran, M., Delay, B., ... & Hamilton, R. B. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', ''67''(3), 883-892. [4] Thorpe, S. G., Thibeault, C. M., Canac, N., Wilk, S. J., Devlin, T., & Hamilton, R. B. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', ''9'', 847. kn93dvghhlk2rtuoo5zph5go1109gky 378 377 2023-12-14T18:59:34Z KedarGrama 6 378 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes the use of Openwater's Gen2 blood flow device which uses Laser Speckle Contrast Imaging (LSCI) to measure relative blood flow in the context of detecting Large Vessel Occlusion (LVO). It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants (see [[link to LVO study]]). ==Feature Extraction and Classification== A large set of features is implemented in the PulseFeatures class, based on published literature ([[1]], [[2]], [[3]]). These features include: * Pulse area under the curve (AUC) [[1]] * Pulse AUC to P1 peak [[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy [[1]] * Pulse onset time [[1]] * Velocity Curvature Index [[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. ==References== * [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. *PloS one*, 15(2), e0228642. * [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. *Translational stroke research*, 10, 475-484. * [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. *IEEE Transactions on Biomedical Engineering*, 67(3), 883-892. pb0ql85wim7gvvcxu1n1nfqlemr3bfo 379 378 2023-12-14T19:06:51Z KedarGrama 6 379 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes the use of Openwater's Gen2 blood flow device which uses Laser Speckle Contrast Imaging (LSCI) to measure relative blood flow in the context of detecting Large Vessel Occlusion (LVO). It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants (see [[link to LVO study]]). ==Feature Extraction and Classification== A large set of features is implemented in the PulseFeatures class, based on published literature ([[1]], [[2]], [[3]]). These features include: * Pulse area under the curve (AUC) [[1]] * Pulse AUC to P1 peak [[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy [[1]] * Pulse onset time [[1]] * Velocity Curvature Index [[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. "PloS one", 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. "Translational stroke research", 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. "IEEE Transactions on Biomedical Engineering", 67(3), 883-892. [[4]] Thorpe, S. G., Thibeault, C. M., Canac, N., Wilk, S. J., Devlin, T., & Hamilton, R. B. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. "Frontiers in neurology", 9, 847. mhmrlpsq007eh6jgjs5lalcu4v3z16g 380 379 2023-12-14T19:07:59Z KedarGrama 6 /* References */ 380 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes the use of Openwater's Gen2 blood flow device which uses Laser Speckle Contrast Imaging (LSCI) to measure relative blood flow in the context of detecting Large Vessel Occlusion (LVO). It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants (see [[link to LVO study]]). ==Feature Extraction and Classification== A large set of features is implemented in the PulseFeatures class, based on published literature ([[1]], [[2]], [[3]]). These features include: * Pulse area under the curve (AUC) [[1]] * Pulse AUC to P1 peak [[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy [[1]] * Pulse onset time [[1]] * Velocity Curvature Index [[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., Thibeault, C. M., Canac, N., Wilk, S. J., Devlin, T., & Hamilton, R. B. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. k2cr4gr9f74gympj97a4937le7ql8nw 381 380 2023-12-14T19:08:58Z KedarGrama 6 /* References */ 381 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes the use of Openwater's Gen2 blood flow device which uses Laser Speckle Contrast Imaging (LSCI) to measure relative blood flow in the context of detecting Large Vessel Occlusion (LVO). It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants (see [[link to LVO study]]). ==Feature Extraction and Classification== A large set of features is implemented in the PulseFeatures class, based on published literature ([[1]], [[2]], [[3]]). These features include: * Pulse area under the curve (AUC) [[1]] * Pulse AUC to P1 peak [[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy [[1]] * Pulse onset time [[1]] * Velocity Curvature Index [[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. k19ok6jm1zu4rh4yhx3e5ruihoctdv0 382 381 2023-12-14T19:10:37Z KedarGrama 6 382 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes the use of Openwater's Gen2 blood flow device which uses Laser Speckle Contrast Imaging (LSCI) to measure relative blood flow in the context of detecting Large Vessel Occlusion (LVO). It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants (see [[link to LVO study]]). ==Feature Extraction and Classification== A large set of features is implemented in the PulseFeatures class, based on published literature ([[1]], [[2]], [[3]]). These features include: * Pulse area under the curve (AUC) [[1]] * Pulse AUC to P1 peak [[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy [[1]] * Pulse onset time [[1]] * Velocity Curvature Index [[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. am3lwq44ytbsflkbhlo7g8tn1r268sk 383 382 2023-12-14T19:11:31Z KedarGrama 6 383 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes the use of Openwater's Gen2 blood flow device which uses Laser Speckle Contrast Imaging (LSCI) to measure relative blood flow in the context of detecting Large Vessel Occlusion (LVO). It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants (see [[link to LVO study]]). ==Feature Extraction and Classification== A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. 1l96hej2kjb8humh7qd92hym98tcdvk 388 383 2023-12-14T21:11:28Z 50.227.118.138 388 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes algorithms used by Openwater's Gen2 blood flow for detecting Large Vessel Occlusion (LVO) stroke. It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. The Openwater system uses near-infrared light to interogate blood flow and blood volume inside the body. For more information on the importance of detecting LVO stroke, the underlying technology we use, and our initial testing, see the [[Blood Flow Technology and Stroke Detection]] wiki. Additional information can be found in our [[Publications]]. In this wiki we decribe the AI algorithms we use to differentiate patients with and without LVO stroke. Our algorithms use machine learning to classify whether subject are likely undergoing a LVO stroke based on the morphology (i.e. shape) of the measured blood flow waveforms (exa. We explored several types of algrorithms accomplish this task, before discovering the one with optimal performance. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants (see ). ==Feature Extraction and Classification== A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. 0vc4se2aak4uo4kxal2kd0x9m8r3vzz 389 388 2023-12-14T21:19:32Z Soren 11 389 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes algorithms used by Openwater's Gen2 blood flow for detecting Large Vessel Occlusion (LVO) stroke. It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. The Openwater system uses near-infrared light to interogate blood flow and blood volume inside the body. For more information on the importance of detecting LVO stroke, the underlying technology we use, and our initial testing, see the [[Blood Flow Technology and Stroke Detection]] wiki. Additional information can be found in our [[Publications]]. In this wiki we decribe the AI algorithms we use to differentiate patients with and without LVO stroke. Our algorithms use machine learning to classify whether subject are likely undergoing a LVO stroke based on the morphology (i.e. shape) of the measured blood flow waveforms (example data below). Several several types of algrorithms were implemented and tested in order to improve performace. These algorithms include random forests, convolutional neural networks, and ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants (see ). ==Feature Extraction and Classification== A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. n7w1kepp3rfuodrukd0xcljdckyernd 391 389 2023-12-14T21:24:53Z Soren 11 391 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes algorithms used by Openwater's Gen2 blood flow for detecting Large Vessel Occlusion (LVO) stroke. It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. The Openwater system uses near-infrared light to interogate blood flow and blood volume inside the body. For more information on the importance of detecting LVO stroke, the underlying technology we use, and our initial testing, see the [[Blood Flow Technology and Stroke Detection]] wiki. Additional information can be found in our [[Publications]]. In this wiki we decribe the AI algorithms we use to differentiate patients with and without LVO stroke. Our algorithms use machine learning to classify whether subject are likely undergoing a LVO stroke based on the morphology (i.e. shape) of the measured blood flow waveforms (example data below). Several several types of algrorithms were implemented and tested in order to improve performace. These algorithms include random forests, deep learning using convolutional neural networks, and deep learning using a transformer architecture. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Near (interrogates superficial flow) * Horizontal (forehead medial) * Vertical (lower temple) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants (see ). == Random Forests == === Feature Extraction and Classification === A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] === Algorithm Description === === Classification Accuracy === Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. == Convolutional Neural Network == === Algorithm Description === === Classification Accuracy === == Transformer Network == === Algorithm Description === === Classification Accuracy === ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. 74rwijiysx04ew32mvfdjz1yljgm826 393 391 2023-12-14T21:33:45Z Soren 11 393 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes algorithms used by Openwater's Gen2 blood flow for detecting Large Vessel Occlusion (LVO) stroke. It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. The Openwater system uses near-infrared light to interogate blood flow and blood volume inside the body. For more information on the importance of detecting LVO stroke, the underlying technology we use, and our initial testing, see the [[Blood Flow Technology and Stroke Detection]] wiki. Additional information can be found in our [[Publications]]. In this wiki we decribe the AI algorithms we use to differentiate patients with and without LVO stroke. Our algorithms use machine learning to classify whether subject are likely undergoing a LVO stroke based on the morphology (i.e. shape) of the measured blood flow waveforms (see example below). Several several types of algrorithms were implemented and tested in order to improve performace. These algorithms include random forests, deep learning using convolutional neural networks, and deep learning using a transformer architecture. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Forehead * Temple These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. [[File:ExampleGen2Data.png|center|575x575px|Representative blood flow waveform data]] == Random Forests == === Feature Extraction and Classification === A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] === Algorithm Description === === Classification Accuracy === Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. == Convolutional Neural Network == === Algorithm Description === === Classification Accuracy === == Transformer Network == === Algorithm Description === === Classification Accuracy === ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. 74zh5g6b5gy6zl5dr3r488t2i514ren 394 393 2023-12-14T21:37:27Z Soren 11 394 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes algorithms used by Openwater's Gen2 blood flow for detecting Large Vessel Occlusion (LVO) stroke. It focuses on the data acquisition, feature extraction, and classification methods employed for this purpose. The Openwater system uses near-infrared light to interogate blood flow and blood volume inside the body. For more information on the importance of detecting LVO stroke, the underlying technology we use, and our initial testing, see the [[Blood Flow Technology and Stroke Detection]] wiki. Additional information can be found in our [[Publications]]. In this wiki we decribe the AI algorithms we use to differentiate patients with and without LVO stroke. Our algorithms use machine learning to classify whether subject are likely undergoing a LVO stroke based on the morphology (i.e. shape) of the measured blood flow waveforms. Several several types of algrorithms were implemented and tested in order to improve performace. These algorithms include random forests, deep learning using convolutional neural networks, and deep learning using a transformer architecture. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Forehead * Temple These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. [[File:ExampleGen2Data.png|center|575x575px|Representative blood flow waveform data]] == Random Forests == === Feature Extraction and Classification === A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] === Algorithm Description === === Classification Accuracy === Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. == Convolutional Neural Network == === Algorithm Description === === Classification Accuracy === == Transformer Network == === Algorithm Description === === Classification Accuracy === ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. cake6msrz4scellvk2fa61tk3l639vv 398 394 2023-12-14T22:44:04Z KedarGrama 6 398 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes algorithms used by Openwater's Gen2 blood flow for detecting Large Vessel Occlusion (LVO) stroke. It focuses on the data acquisition and classification methods employed for this purpose. The Openwater system uses near-infrared light to interrogate blood flow and blood volume inside the body. For more information on the importance of detecting LVO stroke, the underlying technology we use, and our initial testing, see the [[Openwater Stroke Diagnosis Technology]] wiki. Additional information can be found in our [[Publications]]. In this wiki we describe the methods we use to differentiate between patients with and without LVO stroke. We start by based on the morphology (i.e. shape) of the measured blood flow waveforms. We begin by describing the I types of algorithms were implemented and tested in order to improve performance. These algorithms include random forests, deep learning using convolutional neural networks, and deep learning using a transformer architecture. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Forehead * Temple These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. [[File:ExampleGen2Data.png|center|575x575px|Representative blood flow waveform data]] == Random Forests == === Feature Extraction and Classification === A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] === Algorithm Description === === Classification Accuracy === Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. == Convolutional Neural Network == === Algorithm Description === === Classification Accuracy === == Transformer Network == === Algorithm Description === === Classification Accuracy === ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. goq6p19m4j49dpfk8d1f8l50sukyftb 409 398 2023-12-15T00:01:39Z KedarGrama 6 409 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes algorithms used by Openwater's Gen2 blood flow for detecting Large Vessel Occlusion (LVO) stroke. It focuses on the data acquisition and classification methods employed for this purpose. The Openwater system uses near-infrared light to interrogate blood flow and blood volume inside the body. For more information on the importance of detecting LVO stroke, the underlying technology we use, and our initial testing, see the [[Openwater Stroke Diagnosis Technology]] wiki. Additional information can be found in our [[Publications]]. In this wiki we describe the methods we use to differentiate between patients with and without LVO stroke. We started our analysis of the data by based on the morphology (i.e. shape) of the measured blood flow waveforms. We begin by describing the I types of algorithms were implemented and tested in order to improve performance. These algorithms include random forests, deep learning using convolutional neural networks, and deep learning using a transformer architecture. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Forehead * Temple These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. [[File:ExampleGen2Data.png|center|575x575px|Representative blood flow waveform data]] == Random Forests == === Feature Extraction and Classification === A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] === Algorithm Description === === Classification Accuracy === Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script for use with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The classification method achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. == Convolutional Neural Network == === Algorithm Description === === Classification Accuracy === == Transformer Network == === Algorithm Description === === Classification Accuracy === ==References== [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. 895zjlzei2cg9aqudsxa120uraibd0v 410 409 2023-12-15T00:05:54Z KedarGrama 6 /* Random Forests */ 410 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== This article describes algorithms used by Openwater's Gen2 blood flow for detecting Large Vessel Occlusion (LVO) stroke. It focuses on the data acquisition and classification methods employed for this purpose. The Openwater system uses near-infrared light to interrogate blood flow and blood volume inside the body. For more information on the importance of detecting LVO stroke, the underlying technology we use, and our initial testing, see the [[Openwater Stroke Diagnosis Technology]] wiki. Additional information can be found in our [[Publications]]. In this wiki we describe the methods we use to differentiate between patients with and without LVO stroke. We started our analysis of the data by based on the morphology (i.e. shape) of the measured blood flow waveforms. We begin by describing the I types of algorithms were implemented and tested in order to improve performance. These algorithms include random forests, deep learning using convolutional neural networks, and deep learning using a transformer architecture. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Forehead * Temple These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. [[File:ExampleGen2Data.png|center|575x575px|Representative blood flow waveform data]] == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. == LVO Classification with Deep Learning == === Residual Neural Network === === Transformer Network === == Results and Discussion == === References === [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. nncms76xgn57rxsb5a3fzrldydxhhdy 427 410 2023-12-15T03:14:10Z KedarGrama 6 427 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Forehead * Temple These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. [[File:ExampleGen2Data.png|center|575x575px|Representative blood flow waveform data]] == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] as well, it did not improve the classification accuracy. == LVO Classification with Deep Learning == === Residual Neural Network === === Transformer Network === == Results and Discussion == === References === [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. oyc8nj8v18hkroxra917plupgfhxl4x 450 427 2023-12-15T19:04:57Z KedarGrama 6 /* Feature Extraction and LVO Classification */ 450 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Forehead * Temple These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. [[File:ExampleGen2Data.png|center|575x575px|Representative blood flow waveform data]] == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)[[1]] * Pulse AUC to P1 peak[[1]] * Amplitude * Average * Modulation depth (pulsatility index)[[3]] * Skewness * Kurtosis * Pulse canopy[[1]] * Pulse onset time[[1]] * Velocity Curvature Index[[2]] Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index [[4]] <ref>E. Miller, ''The Sun'', (New York: Academic Press, 2005), 23–25.</ref> as well, it did not improve the classification accuracy. == LVO Classification with Deep Learning == === Residual Neural Network === === Transformer Network === == Results and Discussion == === References === [[1]] Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642. [[2]] Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484. [[3]] Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. [[4]] Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847. b4u4utey3rvci35kpu9tjn4yj694i2x 456 450 2023-12-15T19:15:43Z KedarGrama 6 456 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Forehead * Temple These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. [[File:ExampleGen2Data.png|center|575x575px|Representative blood flow waveform data]] == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == === Residual Neural Network === === Transformer Network === == Results and Discussion == == References == <references /> 9hphj6g4sdgkvsmynjf2jryabgwg08f 461 456 2023-12-15T19:25:14Z KedarGrama 6 /* Data Acquisition */ 461 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Forehead * Temple These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths=700px> File: Gen2ContrastandMeanPlot.png|center|Representative blood flow waveform data </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == === Residual Neural Network === === Transformer Network === == Results and Discussion == == References == <references /> 18g3eleb50qtw3wd2sjr38uz1c4r76h 462 461 2023-12-15T19:28:55Z KedarGrama 6 462 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Forehead * Temple These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="1200"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. Black represents the raw speckle contrast data and red the image mean. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == === Residual Neural Network === === Transformer Network === == Results and Discussion == == References == <references /> 7u2op9ri44rj9iyrj18an7wxesfvklx 463 462 2023-12-15T19:29:37Z KedarGrama 6 /* Data Acquisition */ 463 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="1200"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. Black represents the raw speckle contrast data and red the image mean. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == === Residual Neural Network === === Transformer Network === == Results and Discussion == == References == <references /> 6i9wikftgd4iqg508l16i8vpufce58o 478 463 2023-12-15T19:52:46Z KedarGrama 6 /* Data Acquisition */ 478 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. Black represents the raw speckle contrast data and red the image mean. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == === Residual Neural Network === === Transformer Network === == Results and Discussion == == References == <references /> dyq2q0sp1xt4r14hcm4p7ttv23dftme 482 478 2023-12-15T19:55:46Z KedarGrama 6 /* Data Acquisition */ 482 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~68% and specificity of 80%, with an AUC of ~0.72. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == === Residual Neural Network === === Transformer Network === == Results and Discussion == == References == <references /> aueiamcctm17o111oxwjn72fnzeq4uc 491 482 2023-12-15T20:40:53Z KedarGrama 6 491 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == While manually engineered features and robust classifiers like Random forests allow for explainable classifiers, deep neural networks have often overtaken the performance of these methods. We detail a few methods we tried at Openwater with the LVO classification data. === Residual Neural Network(RNN) === We tried two of iterations of RNNs. The first was a 1D RNN <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>. The implementation is available [here](https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py). The model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Since metadata like age, blood pressure and RACE(or LAMS) scores are commonly available to paramedics in the field, we augmented the model to use the metadata when available<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. Adding the metadata boosted the sensitivity to 76% at 80% specificity. This implementation is available [here](https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py). === Transformer Network === == Results and Discussion == == References == <references /> 785womylguo1764soqky9sht1ltz7d8 495 491 2023-12-15T20:53:23Z KedarGrama 6 /* Residual Neural Network(RNN) */ 495 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. Here, we explore two prominent deep learning architectures Openwater has investigated: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Networks === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We used a CNN with residual connections<ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, which enhance performance and learning stability. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. The model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === == Results and Discussion == == References == <references /> izbhurb58fib4cn2e310ckauuprt5i0 520 495 2023-12-15T22:35:27Z KedarGrama 6 /* LVO Classification with Deep Learning */ 520 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. Here, we explore two prominent deep learning architectures Openwater has investigated: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We used a CNN with residual connections<ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, which enhance performance and learning stability. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. We improved the performance of the model by using batch normalization between layers and replacing the ReLU units with GeLU. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have superseded the performance of CNNs in most domains and often perform in learning discriminative representations from biological signals. We improved on a described deep learning model that effectively recognizes ECG waveform abnormalities<ref name="Natarajan2020" />. This model employs a transformer architecture, which effectively extracts distinctive feature representations from the speckle contrast waveform data. We leverage self-attention pooling on the outputs of the transformer layers to enhance the model's performance<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. Like the CNN model, this model could potentially benefit from using incorporating metadata. In our experiments, the benefit was marginal. We speculate that a larger dataset or effective embedding of the metadata might be needed for boosting the accuracy while metadata is available. == Results and Discussion == == References == <references /> 2s05ju82uo1mb9hx0aaqhyt3xe3te0z 521 520 2023-12-15T22:38:57Z KedarGrama 6 521 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We used a CNN with residual connections<ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, which enhance performance and learning stability. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. We improved the performance of the model by using batch normalization between layers and replacing the ReLU units with GeLU. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have superseded the performance of CNNs in most domains and often perform in learning discriminative representations from biological signals. We improved on a described deep learning model that effectively recognizes ECG waveform abnormalities<ref name="Natarajan2020" />. This model employs a transformer architecture, which effectively extracts distinctive feature representations from the speckle contrast waveform data. We leverage self-attention pooling on the outputs of the transformer layers to enhance the model's performance<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. Like the CNN model, this model could potentially benefit from using incorporating metadata. In our experiments, the benefit was marginal. We speculate that a larger dataset or effective embedding of the metadata might be needed for boosting the accuracy while metadata is available. == Results and Discussion == == References == <references /> 119uazxyhslya8bhhdpxeds5mglf4fh 522 521 2023-12-15T22:43:22Z KedarGrama 6 /* Convolutional Neural Network */ 522 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. is model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata == Results and Discussion == == References == <references /> ciu5lva8wns387n6hs4rgwhilr35xqp 532 522 2023-12-15T23:07:20Z KedarGrama 6 /* Results and Discussion */ 532 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. is model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can alleviate unnecessary hospital admissions and resource strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. [[File:Wp figure13bc.png|thumb|ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale.]] == References == <references /> 35bw3wd19e8iwl13uweh34ebbtlqlil 533 532 2023-12-15T23:09:27Z KedarGrama 6 533 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. is model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can alleviate unnecessary hospital admissions and resource strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> 2c5n4z6d2pb28fz7314ggtusqjt4ns2 536 533 2023-12-15T23:12:48Z KedarGrama 6 /* Feature Extraction and LVO Classification */ 536 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File: SampleAveragedBloodflow.png|These plot show </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. is model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can alleviate unnecessary hospital admissions and resource strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> dxqjyap0n29diutyezk41j5dsm7noci 538 536 2023-12-15T23:16:11Z KedarGrama 6 /* Feature Extraction and LVO Classification */ 538 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. is model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can alleviate unnecessary hospital admissions and resource strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> f92fdgrb3f2nslju23f6npc91q9m3w2 542 538 2023-12-15T23:23:34Z 50.227.118.138 542 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. is model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers shortening transit times and alleivating resource strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> 0tff4bh4nhkjw1x3kz55y6bk0ozkhvm 543 542 2023-12-15T23:24:52Z 50.227.118.138 543 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. is model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and lessen resource strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> q3nf5ctg4a4gfk3kjt60pw6xcw9orpy 545 543 2023-12-15T23:26:01Z 50.227.118.138 545 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. is model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> 4hqh58lbs1nlpoydm2ytzth6x007xrn 569 545 2023-12-16T05:24:34Z 135.180.195.174 /* LVO Classification with Deep Learning */ 569 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == .Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. is model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> ewlnfamhh6l0sj6y6gz6uq89v54oh6q 570 569 2023-12-16T05:29:04Z 135.180.195.174 /* LVO Classification with Deep Learning */ 570 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |image = Image of LSCI device |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. This model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> 1pjkualz7vsn8j8drdct1kdduqs4bp9 572 570 2023-12-16T05:51:14Z 135.180.195.174 572 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642.</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892. </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484.</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847.</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining''. 2020. </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. "A wide and deep transformer neural network for 12-lead ECG classification." 2020 Computing in Cardiology. IEEE, 2020. </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. This model, leveraging the power of self-attention pooling<ref>Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences''13.11 (2023): 6410.</ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref>Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184.</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> 4tp82cgwqcb0xni48djrxomi8qvjt46 573 572 2023-12-16T06:05:48Z 135.180.195.174 573 wikitext text/x-wiki {{Infobox |name = Laser Speckle Contrast Imaging (LSCI) for LVO Detection |caption = LSCI device used for measuring relative blood flow. }} ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892 </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. (2020). Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining'' </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. (2020). "A wide and deep transformer neural network for 12-lead ECG classification." ''Computing in Cardiology''. IEEE, 2020 </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. This model, leveraging the power of self-attention pooling<ref> Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences'' 13.11 (2023): 6410 </ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref> Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> jn2uwfyyf40ju3h5hovudwj9db8smty 588 573 2023-12-17T22:57:45Z 73.202.114.83 588 wikitext text/x-wiki ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [[Publications]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892 </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. (2020). Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining'' </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. (2020). "A wide and deep transformer neural network for 12-lead ECG classification." ''Computing in Cardiology''. IEEE, 2020 </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. This model, leveraging the power of self-attention pooling<ref> Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences'' 13.11 (2023): 6410 </ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref> Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> fafl5szcwad1fnd39m0fs70rpsgqkoj 608 588 2023-12-19T01:56:03Z KedarGrama 6 608 wikitext text/x-wiki ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 publication]. All our publications can be found [[Publications|here]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892 </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. (2020). Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining'' </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. (2020). "A wide and deep transformer neural network for 12-lead ECG classification." ''Computing in Cardiology''. IEEE, 2020 </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. This model, leveraging the power of self-attention pooling<ref> Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences'' 13.11 (2023): 6410 </ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref> Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> nrqabl22zsvgpwb74ycky1ki4qogppf 609 608 2023-12-19T01:56:20Z KedarGrama 6 609 wikitext text/x-wiki ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 publication]. A list of our publications can be found [[Publications|here]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892 </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. (2020). Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining'' </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. (2020). "A wide and deep transformer neural network for 12-lead ECG classification." ''Computing in Cardiology''. IEEE, 2020 </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. This model, leveraging the power of self-attention pooling<ref> Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences'' 13.11 (2023): 6410 </ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref> Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> q05q2pidjbit4nwy6jwcfxboantndh2 610 609 2023-12-19T01:57:05Z KedarGrama 6 610 wikitext text/x-wiki ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 publication] from the clinical study. A list of our publications can be found [[Publications|here]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892 </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. (2020). Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining'' </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. (2020). "A wide and deep transformer neural network for 12-lead ECG classification." ''Computing in Cardiology''. IEEE, 2020 </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. This model, leveraging the power of self-attention pooling<ref> Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences'' 13.11 (2023): 6410 </ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref> Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> 1bpw04e72hf9t28dqxquovqvs18nsib 623 610 2023-12-19T18:38:58Z Admin 1 Protected "[[Blood Flow Gen 2 LVO Classification and Analysis]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 610 wikitext text/x-wiki ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 publication] from the clinical study. A list of our publications can be found [[Publications|here]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892 </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. (2020). Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining'' </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. (2020). "A wide and deep transformer neural network for 12-lead ECG classification." ''Computing in Cardiology''. IEEE, 2020 </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. This model, leveraging the power of self-attention pooling<ref> Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences'' 13.11 (2023): 6410 </ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref> Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> 1bpw04e72hf9t28dqxquovqvs18nsib 639 623 2023-12-19T19:45:46Z KedarGrama 6 /* Transformer Network */ 639 wikitext text/x-wiki ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 publication] from the clinical study. A list of our publications can be found [[Publications|here]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892 </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. (2020). Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining'' </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. (2020). "A wide and deep transformer neural network for 12-lead ECG classification." ''Computing in Cardiology''. IEEE, 2020 </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. This model, leveraging the power of self-attention pooling<ref> Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences'' 13.11 (2023): 6410 </ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref> Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We experimented with adding the metadata as final fully connected layer inputs<ref name="Natarajan2020" /> and as inputs to the transformer with separate positional encodings. Neither provided significant improvement to the classification. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> lt6w5ealqnwdbsdvie0dercbou632tm 671 639 2024-01-03T00:54:15Z KedarGrama 6 671 wikitext text/x-wiki ==Introduction== Openwater's Gen2 blood flow leverages innovative algorithms to detect Large Vessel Occlusion (LVO) strokes, focusing on classification techniques. Source code for all the methods described are available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai here]. The system utilizes near-infrared light to analyze blood flow and volume within the body. For context on LVO stroke's critical role, the underlying technology, and initial testing results, please refer to the [[Openwater Stroke Diagnosis Technology]] wiki and our [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 publication] from the clinical study. A list of our publications can be found [[Publications|here]]. This wiki delves into the specific methods we employ to differentiate LVO stroke patients from those without. Our journey to classify the data began with manual feature extraction from measured blood flow waveforms. We detail the various algorithms implemented and tested to optimize performance, including random forests, convolutional neural network-powered deep learning, and transformer-based deep learning approaches. ==Data Acquisition== The device provides relative blood flow measurements at three forehead positions: * Surface (interrogates superficial flow) * Vertical (Temple) * Horizontal (Forehead - Medial) These positions roughly map to the blood flow supplied by the internal carotid artery (ICA) and middle cerebral artery (MCA). Data acquisition involves recording 15 seconds of simultaneous blood flow data on both sides of the head at each position. The acquisition order follows a specific pattern for most LVO study participants. <gallery widths="800" heights="259"> File:Gen2ContrastandMeanPlot.png|Representative blood flow waveform data. The black plot shows the raw speckle contrast, capturing fluctuations due to blood movement. The red plot represents the image mean, indicating the average intensity across the image which corresponds to the volume of flow. </gallery> == Feature Extraction and LVO Classification == A large set of features is implemented in the PulseFeatures class, based on published literature. These features include: * Pulse area under the curve (AUC)<ref name="Thorpe2020"> Thorpe, S. G., et al. (2020). Toward automated classification of pathological transcranial Doppler waveform morphology via spectral clustering. ''PloS one'', 15(2), e0228642</ref> * Pulse AUC to P1 peak<ref name="Thorpe2020" /> * Amplitude * Average * Modulation depth (pulsatility index)<ref name="Jalaleddini2019"> Jalaleddini, K., et al. (2019). Objective assessment of beat quality in transcranial Doppler measurement of blood flow velocity in cerebral arteries. ''IEEE Transactions on Biomedical Engineering'', 67(3), 883-892 </ref> * Skewness * Kurtosis * Pulse canopy<ref name="Thorpe2020" /> * Pulse onset time<ref name="Thorpe2020" /> * Velocity Curvature Index<ref name="Thorpe2019"> Thorpe, S. G., et al. (2019). Velocity curvature index: a novel diagnostic biomarker for large vessel occlusion. ''Translational stroke research'', 10, 475-484</ref> Feature calculation is performed on the average pulse in a scan and each individual pulse in a scan (median and range used as features for classification). This feature extraction approach is implemented in the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/runFeatureAnalysis.py runFeatureAnalysis] script and used with a random forest classifier. The raw features are implemented in [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/ReadGen2Data.py#L223 PulseFeatures]. The Random Forest classification achieves a sensitivity of ~55% and specificity of 80%, with an AUC of ~0.78. This performance is comparable to existing LVO detection methods like RACE and LAMS. We experimented with Velocity Asymmetry Index<ref name="Thorpe2018"> Thorpe, S. G., et al. (2018). Decision criteria for large vessel occlusion using transcranial Doppler waveform morphology. ''Frontiers in neurology'', 9, 847</ref> as well, it did not improve the LVO classification accuracy with data from our device. <gallery widths="534" heights="493"> File:SampleAveragedBloodflow.png|The gray plots show the speckle contrast waveform for each pulse in a 15 second acquisition of data on the left horizontal sensor in two subjects, one with an LVO, and the other, a healthy volunteer. </gallery> == LVO Classification with Deep Learning == Deep learning has emerged as a powerful tool for LVO classification, potentially surpassing the performance of traditional methods like Random Forests. At Openwater, we've explored two prominent deep learning architectures: Convolutional Neural Networks (CNNs) and Transformer Networks. === Convolutional Neural Network === CNNs excel at processing data with inherent spatial patterns, making them ideal for analyzing blood flow waveforms. We leveraged a CNN with residual connections <ref name="hong2020"> Hong, Shenda, et al. (2020). Holmes: health online model ensemble serving for deep learning models in intensive care units. ''Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining'' </ref>, enhancing its performance and learning stability by using batch normalization between layers and replacing the ReLU units with GeLU. The base implementation is available [https://github.com/hsd1503/resnet1d/blob/9cccb9f08fde65087272472d0572c2dbe16a8119/resnet1d.py here]. Our model achieves a sensitivity of 73% at 80% specificity with an AUC of 0.77. Deep learning models often benefit from data from multiple domains. Recognizing the availability of metadata like age, blood pressure, and RACE scores for paramedics, we augmented our model to utilize this information when available. This is implemented as an additional fully connected layer in the final layer of the CNN<ref name="Natarajan2020"> Natarajan, Annamalai, et al. (2020). "A wide and deep transformer neural network for 12-lead ECG classification." ''Computing in Cardiology''. IEEE, 2020 </ref>. This simple addition boosted the sensitivity of our CNN model to 76% at 80% specificity, demonstrating the potential of incorporating additional data sources. Implementation is available [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai/blob/f19cded9e80c14f35ce74388fb27c375210e0b11/DeepLearning/resnet1d.py here]. === Transformer Network === Transformer networks have emerged as the new champions of representation learning across various domains. Inspired by their success, we fine-tuned a transformer model originally designed for ECG waveform analysis<ref name="Natarajan2020" />. This model, leveraging the power of self-attention pooling<ref> Safari, Pooyan, Miquel India, and Javier Hernando. "Self Attention Networks in Speaker Recognition." ''Applied Sciences'' 13.11 (2023): 6410 </ref>, effectively extracts distinctive features from the complex speckle contrast waveform data. The network's output is then converted into a probability score for either the LVO or the non-LVO class using the SoftMax function<ref> Courville A. 6.2.2.3 Softmax Units for Multinoulli Output Distributions. In: ''Deep Learning''. MIT Press; 2016:180-184</ref>. Results of the classification are described below. The results, while promising, hinted at a potential barrier to unlocking the full potential of metadata in this context. We experimented with adding the metadata as final fully connected layer inputs<ref name="Natarajan2020" /> and as inputs to the transformer with separate positional encodings. Neither provided significant improvement to the classification. We hypothesize that a larger dataset or a more sophisticated embedding technique might be necessary to fully unleash the synergy between transformers and metadata. == Results and Discussion == In a [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 clinical study] conducted at two comprehensive stroke centers, the Openwater headset demonstrated superior performance compared to traditional prehospital stroke scales for detecting Large Vessel Occlusions (LVOs) in patients undergoing acute stroke evaluation. The device achieved significantly higher sensitivity, specificity, and area under the curve (AUC) compared to established scales, suggesting its potential to improve stroke diagnosis accuracy in the critical early stages. {| class="wikitable" |+Sensitivity and specificity comparison between diagnostic tools. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. ! colspan="3" |Diagnostic Performance N=135 ! colspan="4" |Based on 5% LVO prevalence in a sample of n=1000 ! colspan="4" |Based on 10% LVO prevalence in a sample of n=1000 |- | |Sensitivity |Specificity |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |Positive Predictive Value |Negative Predictive Value |Number of False Positives |Number of False Negatives |- |Openwater |78.9% |84.3% |20.9% |98.7% |149 |11 |35.8% |97.3% |141 |16 |- |RACE |59.6% |80.7% |14% |97.4% |183 |20 |25.5% |94.7% |174 |40 |- |LAMS |50% |80.7 |12% |96.8% |183 |25 |22.4% |93.6% |174 |50 |} While these results are promising, further research is needed to evaluate the Openwater headset's performance in prehospital environments. Our clinical study where we had 135 successful scans on patients with 52 of them being LVOs and 83 other non-LVO strokes and mimics. A typical Emergency Medical Services (EMS) system encounters a lower prevalence of LVOs (around 5-10%) compared to the study setting. To ensure optimal performance in real-world scenarios, we are actively investigating the device's accuracy in these conditions. Benefits of Improved LVO Detection: * Reduced False Positives: Lowering false positives can reduce unnecessary routing to comprehensive stroke centers, shortening transit times and alleviating strain on the medical system. * Reduced False Negatives: Timely identification of LVOs allows for faster initiation of critical reperfusion therapy, significantly improving patient outcomes. <gallery widths="800" heights="485"> File: Wp figure13bc.png| ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. </gallery> == References == <references /> tn45jts59nmiul03l4hn7prdvzsahns 0 34 126 2023-12-13T03:06:36Z Gvigelet 4 Created page with "<span id="openwater-blood-flow-analysis-device---software-architecture-overview"></span> = Openwater Blood Flow Analysis Device - Software Architecture Overview = <span id="purpose"></span> == Purpose == This document provides an overview of the software architecture for the Openwater Blood Flow Analysis device, focusing on its key components and interactions. <span id="architecture-design"></span> == Architecture Design == The software architecture follows a layered..." 126 wikitext text/x-wiki <span id="openwater-blood-flow-analysis-device---software-architecture-overview"></span> = Openwater Blood Flow Analysis Device - Software Architecture Overview = <span id="purpose"></span> == Purpose == This document provides an overview of the software architecture for the Openwater Blood Flow Analysis device, focusing on its key components and interactions. <span id="architecture-design"></span> == Architecture Design == The software architecture follows a layered approach: * '''Presentation Layer''': User interface for interaction with the device. * '''Business Logic Layer''': Core functionality including data processing and analysis. * '''Data Access Layer''': Manages data storage and retrieval. <span id="main-components"></span> == Main Components == * '''User Interaction Module''': Handles user inputs and displays results. * '''Data Management Module''': Responsible for data storage and management. * '''Processing &amp; Analysis Module''': Processes blood flow data and generates analysis. * '''Device Control Module''': Manages device operations and settings. * '''Safety &amp; Alert Module''': Ensures device safety and alerts users of any issues. <span id="diagrams"></span> == Diagrams == High Level Software Component Diagram [[File:ow_system_diagram.png|thumb|none|alt=Openwater System Diagram|Openwater System Diagram]] High Level Data Flow Diagram [[File:ow_dataflow_diagram.png|thumb|none|alt=Openwater Data Flow Diagram|Openwater Data Flow Diagram]] <span id="use-cases"></span> == Use Cases == * Initiating blood flow analysis. * Viewing analysis results. * Adjusting device settings. * Responding to safety alerts. * Exporting data for external review. <span id="constraints"></span> == Constraints == * Hardware limitations. * Regulatory compliance. <span id="goals"></span> == Goals == * Reliability, usability, safety, scalability, interoperability, maintainability, security, and performance. r56jgwvaohiakbhwvm8tcmof3xproeq 129 126 2023-12-13T03:12:34Z Gvigelet 4 /* Diagrams */ 129 wikitext text/x-wiki <span id="openwater-blood-flow-analysis-device---software-architecture-overview"></span> = Openwater Blood Flow Analysis Device - Software Architecture Overview = <span id="purpose"></span> == Purpose == This document provides an overview of the software architecture for the Openwater Blood Flow Analysis device, focusing on its key components and interactions. <span id="architecture-design"></span> == Architecture Design == The software architecture follows a layered approach: * '''Presentation Layer''': User interface for interaction with the device. * '''Business Logic Layer''': Core functionality including data processing and analysis. * '''Data Access Layer''': Manages data storage and retrieval. <span id="main-components"></span> == Main Components == * '''User Interaction Module''': Handles user inputs and displays results. * '''Data Management Module''': Responsible for data storage and management. * '''Processing &amp; Analysis Module''': Processes blood flow data and generates analysis. * '''Device Control Module''': Manages device operations and settings. * '''Safety &amp; Alert Module''': Ensures device safety and alerts users of any issues. <span id="diagrams"></span> == Diagrams == High Level Software Component Diagram <p align="center"> [[File:ow_system_diagram.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure: Openwater System Diagram|Openwater System Diagram</pre> </div> High Level Data Flow Diagram <p align="center"> [[File:ow_dataflow_diagram.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure: Openwater Data Flow Diagram</pre> </div> <span id="use-cases"></span> == Use Cases == * Initiating blood flow analysis. * Viewing analysis results. * Adjusting device settings. * Responding to safety alerts. * Exporting data for external review. <span id="constraints"></span> == Constraints == * Hardware limitations. * Regulatory compliance. <span id="goals"></span> == Goals == * Reliability, usability, safety, scalability, interoperability, maintainability, security, and performance. 195mghwnhumrrsayazszi29qnudpdld 130 129 2023-12-13T03:12:59Z Gvigelet 4 130 wikitext text/x-wiki <span id="openwater-blood-flow-analysis-device---software-architecture-overview"></span> = Openwater Blood Flow Analysis Device - Software Architecture Overview = <span id="purpose"></span> == Purpose == This document provides an overview of the software architecture for the Openwater Blood Flow Analysis device, focusing on its key components and interactions. <span id="architecture-design"></span> == Architecture Design == The software architecture follows a layered approach: * '''Presentation Layer''': User interface for interaction with the device. * '''Business Logic Layer''': Core functionality including data processing and analysis. * '''Data Access Layer''': Manages data storage and retrieval. <span id="main-components"></span> == Main Components == * '''User Interaction Module''': Handles user inputs and displays results. * '''Data Management Module''': Responsible for data storage and management. * '''Processing &amp; Analysis Module''': Processes blood flow data and generates analysis. * '''Device Control Module''': Manages device operations and settings. * '''Safety &amp; Alert Module''': Ensures device safety and alerts users of any issues. <span id="diagrams"></span> == Diagrams == High Level Software Component Diagram <p align="center"> [[File:ow_system_diagram.png|756x756px]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure: Openwater System Diagram|Openwater System Diagram</pre> </div> High Level Data Flow Diagram <p align="center"> [[File:ow_dataflow_diagram.png|781x781px]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure: Openwater Data Flow Diagram</pre> </div> <span id="use-cases"></span> == Use Cases == * Initiating blood flow analysis. * Viewing analysis results. * Adjusting device settings. * Responding to safety alerts. * Exporting data for external review. <span id="constraints"></span> == Constraints == * Hardware limitations. * Regulatory compliance. <span id="goals"></span> == Goals == * Reliability, usability, safety, scalability, interoperability, maintainability, security, and performance. g3rnkw1krxjhd6krnboz5yjvrow39ar 132 130 2023-12-13T03:20:50Z Gvigelet 4 132 wikitext text/x-wiki = Openwater Gen2 Blood Flow Device Overview = == Introduction == The **Openwater Blood Flow Analysis Device** is an innovative diagnostic tool designed to non-invasively measure and monitor cerebral blood flow. Utilizing advanced near-infrared light techniques, it facilitates the detection of hemodynamic changes associated with various medical conditions, particularly cerebrovascular events like strokes. == Fundamental Theory == The device employs near-infrared spectroscopy (NIRS) to penetrate the skull and tissue, allowing for the visualization of blood flow within the brain. The key principles include: - **Near-Infrared Light Penetration**: NIR light, due to its longer wavelength, can traverse the human skull and reach the brain, providing a window into cerebral circulation. - **Hemodynamic Response Detection**: The device measures the dynamic changes in blood oxygenation and volume, which are indicative of the brain's metabolic activity. == Core Technologies == The device integrates several technological advancements to achieve precise blood flow imaging: - **Diffuse Optical Tomography (DOT)**: Generates images by detecting light absorption and scattering patterns, which vary with blood flow and oxygenation. - **Laser Speckle Contrast Imaging (LSCI)**: Measures blood flow velocity by analyzing the speckle pattern of reflected light from moving blood cells. == Device Components and Operation == The operational mechanism of the device is centered around a head mounted senor module that emits NIR light and collects returning signals: - **Sensor Module**: A component that directs NIR light onto the patient's head and captures the backscattered light. - **Processing Core**: Analyzes the captured data using algorithms to map blood flow and oxygenation levels. == Clinical Application == The primary clinical application of this device is the rapid assessment of blood flow in the brain for: - **Stroke Diagnosis**: Quick identification of ischemic regions where blood flow is compromised. - **Monitoring Therapy**: Assessing the efficacy of interventions aimed at restoring normal blood flow. == Advantages Over Traditional Methods == The Openwater Blood Flow Analysis Device offers several benefits: - **Non-invasive Operation**: Unlike invasive cerebral angiography, it poses no risks associated with surgical procedures. - **Accessibility**: Its portability makes it suitable for bedside examinations and potential field use in emergency settings. - **Real-time Analysis**: Provides immediate feedback on cerebral blood flow, crucial for urgent medical decision-making. - **Speed of Testing**: Delivers diagnostic results in less significantly less time compared to traditional imaging methods like MRI, which can take much longer. == Safety Considerations == The device is designed with patient safety in mind and adheres to strict regulatory standards for medical devices. It includes fail-safes and alert systems to prevent any potential risks during operation. == Conclusion == The Openwater Blood Flow Analysis Device represents a significant innovation in medical diagnostics, offering a safe, efficient, and non-invasive method to evaluate cerebral blood flow. It has the potential to be a game-changer in the rapid detection and management of strokes and other conditions affecting cerebral hemodynamics. 7jv7upe15f1r2jatszkhbcd98s4fmye 289 132 2023-12-13T21:54:26Z Gvigelet 4 /* Fundamental Theory */ 289 wikitext text/x-wiki = Openwater Gen2 Blood Flow Device Overview = == Introduction == The **Openwater Blood Flow Analysis Device** is an innovative diagnostic tool designed to non-invasively measure and monitor cerebral blood flow. Utilizing advanced near-infrared light techniques, it facilitates the detection of hemodynamic changes associated with various medical conditions, particularly cerebrovascular events like strokes. == Fundamental Theory == The device employs near-infrared spectroscopy (NIRS) to penetrate the skull and tissue, allowing for the visualization of blood flow within the brain. The key principles include: * '''Near-Infrared Light Penetration''': NIR light, due to its longer wavelength, can traverse the human skull and reach the brain, providing a window into cerebral circulation. * '''Hemodynamic Response Detection''': The device measures the dynamic changes in blood oxygenation and volume, which are indicative of the brain's metabolic activity. == Core Technologies == The device integrates several technological advancements to achieve precise blood flow imaging: - **Diffuse Optical Tomography (DOT)**: Generates images by detecting light absorption and scattering patterns, which vary with blood flow and oxygenation. - **Laser Speckle Contrast Imaging (LSCI)**: Measures blood flow velocity by analyzing the speckle pattern of reflected light from moving blood cells. == Device Components and Operation == The operational mechanism of the device is centered around a head mounted senor module that emits NIR light and collects returning signals: - **Sensor Module**: A component that directs NIR light onto the patient's head and captures the backscattered light. - **Processing Core**: Analyzes the captured data using algorithms to map blood flow and oxygenation levels. == Clinical Application == The primary clinical application of this device is the rapid assessment of blood flow in the brain for: - **Stroke Diagnosis**: Quick identification of ischemic regions where blood flow is compromised. - **Monitoring Therapy**: Assessing the efficacy of interventions aimed at restoring normal blood flow. == Advantages Over Traditional Methods == The Openwater Blood Flow Analysis Device offers several benefits: - **Non-invasive Operation**: Unlike invasive cerebral angiography, it poses no risks associated with surgical procedures. - **Accessibility**: Its portability makes it suitable for bedside examinations and potential field use in emergency settings. - **Real-time Analysis**: Provides immediate feedback on cerebral blood flow, crucial for urgent medical decision-making. - **Speed of Testing**: Delivers diagnostic results in less significantly less time compared to traditional imaging methods like MRI, which can take much longer. == Safety Considerations == The device is designed with patient safety in mind and adheres to strict regulatory standards for medical devices. It includes fail-safes and alert systems to prevent any potential risks during operation. == Conclusion == The Openwater Blood Flow Analysis Device represents a significant innovation in medical diagnostics, offering a safe, efficient, and non-invasive method to evaluate cerebral blood flow. It has the potential to be a game-changer in the rapid detection and management of strokes and other conditions affecting cerebral hemodynamics. rmmt31lfev3mjyy4t3zv9cf6z5gth67 290 289 2023-12-13T21:55:13Z Gvigelet 4 /* Core Technologies */ 290 wikitext text/x-wiki = Openwater Gen2 Blood Flow Device Overview = == Introduction == The **Openwater Blood Flow Analysis Device** is an innovative diagnostic tool designed to non-invasively measure and monitor cerebral blood flow. Utilizing advanced near-infrared light techniques, it facilitates the detection of hemodynamic changes associated with various medical conditions, particularly cerebrovascular events like strokes. == Fundamental Theory == The device employs near-infrared spectroscopy (NIRS) to penetrate the skull and tissue, allowing for the visualization of blood flow within the brain. The key principles include: * '''Near-Infrared Light Penetration''': NIR light, due to its longer wavelength, can traverse the human skull and reach the brain, providing a window into cerebral circulation. * '''Hemodynamic Response Detection''': The device measures the dynamic changes in blood oxygenation and volume, which are indicative of the brain's metabolic activity. == Core Technologies == The device integrates several technological advancements to achieve precise blood flow imaging: * '''Diffuse Optical Tomography (DOT)''': Generates images by detecting light absorption and scattering patterns, which vary with blood flow and oxygenation. * '''Laser Speckle Contrast Imaging (LSCI)''': Measures blood flow velocity by analyzing the speckle pattern of reflected light from moving blood cells. == Device Components and Operation == The operational mechanism of the device is centered around a head mounted senor module that emits NIR light and collects returning signals: - **Sensor Module**: A component that directs NIR light onto the patient's head and captures the backscattered light. - **Processing Core**: Analyzes the captured data using algorithms to map blood flow and oxygenation levels. == Clinical Application == The primary clinical application of this device is the rapid assessment of blood flow in the brain for: - **Stroke Diagnosis**: Quick identification of ischemic regions where blood flow is compromised. - **Monitoring Therapy**: Assessing the efficacy of interventions aimed at restoring normal blood flow. == Advantages Over Traditional Methods == The Openwater Blood Flow Analysis Device offers several benefits: - **Non-invasive Operation**: Unlike invasive cerebral angiography, it poses no risks associated with surgical procedures. - **Accessibility**: Its portability makes it suitable for bedside examinations and potential field use in emergency settings. - **Real-time Analysis**: Provides immediate feedback on cerebral blood flow, crucial for urgent medical decision-making. - **Speed of Testing**: Delivers diagnostic results in less significantly less time compared to traditional imaging methods like MRI, which can take much longer. == Safety Considerations == The device is designed with patient safety in mind and adheres to strict regulatory standards for medical devices. It includes fail-safes and alert systems to prevent any potential risks during operation. == Conclusion == The Openwater Blood Flow Analysis Device represents a significant innovation in medical diagnostics, offering a safe, efficient, and non-invasive method to evaluate cerebral blood flow. It has the potential to be a game-changer in the rapid detection and management of strokes and other conditions affecting cerebral hemodynamics. lqaoi671qn3ra8f0q5u6gkh4w4hnc22 291 290 2023-12-13T21:55:43Z Gvigelet 4 /* Device Components and Operation */ 291 wikitext text/x-wiki = Openwater Gen2 Blood Flow Device Overview = == Introduction == The **Openwater Blood Flow Analysis Device** is an innovative diagnostic tool designed to non-invasively measure and monitor cerebral blood flow. Utilizing advanced near-infrared light techniques, it facilitates the detection of hemodynamic changes associated with various medical conditions, particularly cerebrovascular events like strokes. == Fundamental Theory == The device employs near-infrared spectroscopy (NIRS) to penetrate the skull and tissue, allowing for the visualization of blood flow within the brain. The key principles include: * '''Near-Infrared Light Penetration''': NIR light, due to its longer wavelength, can traverse the human skull and reach the brain, providing a window into cerebral circulation. * '''Hemodynamic Response Detection''': The device measures the dynamic changes in blood oxygenation and volume, which are indicative of the brain's metabolic activity. == Core Technologies == The device integrates several technological advancements to achieve precise blood flow imaging: * '''Diffuse Optical Tomography (DOT)''': Generates images by detecting light absorption and scattering patterns, which vary with blood flow and oxygenation. * '''Laser Speckle Contrast Imaging (LSCI)''': Measures blood flow velocity by analyzing the speckle pattern of reflected light from moving blood cells. == Device Components and Operation == The operational mechanism of the device is centered around a head mounted senor module that emits NIR light and collects returning signals: * '''Sensor Module''': A component that directs NIR light onto the patient's head and captures the backscattered light. * '''Processing Core''': Analyzes the captured data using algorithms to map blood flow and oxygenation levels. == Clinical Application == The primary clinical application of this device is the rapid assessment of blood flow in the brain for: - **Stroke Diagnosis**: Quick identification of ischemic regions where blood flow is compromised. - **Monitoring Therapy**: Assessing the efficacy of interventions aimed at restoring normal blood flow. == Advantages Over Traditional Methods == The Openwater Blood Flow Analysis Device offers several benefits: - **Non-invasive Operation**: Unlike invasive cerebral angiography, it poses no risks associated with surgical procedures. - **Accessibility**: Its portability makes it suitable for bedside examinations and potential field use in emergency settings. - **Real-time Analysis**: Provides immediate feedback on cerebral blood flow, crucial for urgent medical decision-making. - **Speed of Testing**: Delivers diagnostic results in less significantly less time compared to traditional imaging methods like MRI, which can take much longer. == Safety Considerations == The device is designed with patient safety in mind and adheres to strict regulatory standards for medical devices. It includes fail-safes and alert systems to prevent any potential risks during operation. == Conclusion == The Openwater Blood Flow Analysis Device represents a significant innovation in medical diagnostics, offering a safe, efficient, and non-invasive method to evaluate cerebral blood flow. It has the potential to be a game-changer in the rapid detection and management of strokes and other conditions affecting cerebral hemodynamics. fsdabt3pz4y7ouh693nwspc7qooys48 292 291 2023-12-13T21:56:03Z Gvigelet 4 /* Clinical Application */ 292 wikitext text/x-wiki = Openwater Gen2 Blood Flow Device Overview = == Introduction == The **Openwater Blood Flow Analysis Device** is an innovative diagnostic tool designed to non-invasively measure and monitor cerebral blood flow. Utilizing advanced near-infrared light techniques, it facilitates the detection of hemodynamic changes associated with various medical conditions, particularly cerebrovascular events like strokes. == Fundamental Theory == The device employs near-infrared spectroscopy (NIRS) to penetrate the skull and tissue, allowing for the visualization of blood flow within the brain. The key principles include: * '''Near-Infrared Light Penetration''': NIR light, due to its longer wavelength, can traverse the human skull and reach the brain, providing a window into cerebral circulation. * '''Hemodynamic Response Detection''': The device measures the dynamic changes in blood oxygenation and volume, which are indicative of the brain's metabolic activity. == Core Technologies == The device integrates several technological advancements to achieve precise blood flow imaging: * '''Diffuse Optical Tomography (DOT)''': Generates images by detecting light absorption and scattering patterns, which vary with blood flow and oxygenation. * '''Laser Speckle Contrast Imaging (LSCI)''': Measures blood flow velocity by analyzing the speckle pattern of reflected light from moving blood cells. == Device Components and Operation == The operational mechanism of the device is centered around a head mounted senor module that emits NIR light and collects returning signals: * '''Sensor Module''': A component that directs NIR light onto the patient's head and captures the backscattered light. * '''Processing Core''': Analyzes the captured data using algorithms to map blood flow and oxygenation levels. == Clinical Application == The primary clinical application of this device is the rapid assessment of blood flow in the brain for: * '''Stroke Diagnosis''': Quick identification of ischemic regions where blood flow is compromised. * '''Monitoring Therapy''': Assessing the efficacy of interventions aimed at restoring normal blood flow. == Advantages Over Traditional Methods == The Openwater Blood Flow Analysis Device offers several benefits: - **Non-invasive Operation**: Unlike invasive cerebral angiography, it poses no risks associated with surgical procedures. - **Accessibility**: Its portability makes it suitable for bedside examinations and potential field use in emergency settings. - **Real-time Analysis**: Provides immediate feedback on cerebral blood flow, crucial for urgent medical decision-making. - **Speed of Testing**: Delivers diagnostic results in less significantly less time compared to traditional imaging methods like MRI, which can take much longer. == Safety Considerations == The device is designed with patient safety in mind and adheres to strict regulatory standards for medical devices. It includes fail-safes and alert systems to prevent any potential risks during operation. == Conclusion == The Openwater Blood Flow Analysis Device represents a significant innovation in medical diagnostics, offering a safe, efficient, and non-invasive method to evaluate cerebral blood flow. It has the potential to be a game-changer in the rapid detection and management of strokes and other conditions affecting cerebral hemodynamics. alh6ouu7h95viylnq98x757df2231sl 293 292 2023-12-13T21:56:32Z Gvigelet 4 /* Advantages Over Traditional Methods */ 293 wikitext text/x-wiki = Openwater Gen2 Blood Flow Device Overview = == Introduction == The **Openwater Blood Flow Analysis Device** is an innovative diagnostic tool designed to non-invasively measure and monitor cerebral blood flow. Utilizing advanced near-infrared light techniques, it facilitates the detection of hemodynamic changes associated with various medical conditions, particularly cerebrovascular events like strokes. == Fundamental Theory == The device employs near-infrared spectroscopy (NIRS) to penetrate the skull and tissue, allowing for the visualization of blood flow within the brain. The key principles include: * '''Near-Infrared Light Penetration''': NIR light, due to its longer wavelength, can traverse the human skull and reach the brain, providing a window into cerebral circulation. * '''Hemodynamic Response Detection''': The device measures the dynamic changes in blood oxygenation and volume, which are indicative of the brain's metabolic activity. == Core Technologies == The device integrates several technological advancements to achieve precise blood flow imaging: * '''Diffuse Optical Tomography (DOT)''': Generates images by detecting light absorption and scattering patterns, which vary with blood flow and oxygenation. * '''Laser Speckle Contrast Imaging (LSCI)''': Measures blood flow velocity by analyzing the speckle pattern of reflected light from moving blood cells. == Device Components and Operation == The operational mechanism of the device is centered around a head mounted senor module that emits NIR light and collects returning signals: * '''Sensor Module''': A component that directs NIR light onto the patient's head and captures the backscattered light. * '''Processing Core''': Analyzes the captured data using algorithms to map blood flow and oxygenation levels. == Clinical Application == The primary clinical application of this device is the rapid assessment of blood flow in the brain for: * '''Stroke Diagnosis''': Quick identification of ischemic regions where blood flow is compromised. * '''Monitoring Therapy''': Assessing the efficacy of interventions aimed at restoring normal blood flow. == Advantages Over Traditional Methods == The Openwater Blood Flow Analysis Device offers several benefits: * '''Non-invasive Operation''': Unlike invasive cerebral angiography, it poses no risks associated with surgical procedures. * '''Accessibility''': Its portability makes it suitable for bedside examinations and potential field use in emergency settings. * '''Real-time Analysis''': Provides immediate feedback on cerebral blood flow, crucial for urgent medical decision-making. * '''Speed of Testing''': Delivers diagnostic results in less significantly less time compared to traditional imaging methods like MRI, which can take much longer. == Safety Considerations == The device is designed with patient safety in mind and adheres to strict regulatory standards for medical devices. It includes fail-safes and alert systems to prevent any potential risks during operation. == Conclusion == The Openwater Blood Flow Analysis Device represents a significant innovation in medical diagnostics, offering a safe, efficient, and non-invasive method to evaluate cerebral blood flow. It has the potential to be a game-changer in the rapid detection and management of strokes and other conditions affecting cerebral hemodynamics. axhknrav9ot7djngftvshseugkmccpx 294 293 2023-12-13T21:56:54Z Gvigelet 4 /* Introduction */ 294 wikitext text/x-wiki = Openwater Gen2 Blood Flow Device Overview = == Introduction == The '''Openwater Blood Flow Analysis Device''' is an innovative diagnostic tool designed to non-invasively measure and monitor cerebral blood flow. Utilizing advanced near-infrared light techniques, it facilitates the detection of hemodynamic changes associated with various medical conditions, particularly cerebrovascular events like strokes. == Fundamental Theory == The device employs near-infrared spectroscopy (NIRS) to penetrate the skull and tissue, allowing for the visualization of blood flow within the brain. The key principles include: * '''Near-Infrared Light Penetration''': NIR light, due to its longer wavelength, can traverse the human skull and reach the brain, providing a window into cerebral circulation. * '''Hemodynamic Response Detection''': The device measures the dynamic changes in blood oxygenation and volume, which are indicative of the brain's metabolic activity. == Core Technologies == The device integrates several technological advancements to achieve precise blood flow imaging: * '''Diffuse Optical Tomography (DOT)''': Generates images by detecting light absorption and scattering patterns, which vary with blood flow and oxygenation. * '''Laser Speckle Contrast Imaging (LSCI)''': Measures blood flow velocity by analyzing the speckle pattern of reflected light from moving blood cells. == Device Components and Operation == The operational mechanism of the device is centered around a head mounted senor module that emits NIR light and collects returning signals: * '''Sensor Module''': A component that directs NIR light onto the patient's head and captures the backscattered light. * '''Processing Core''': Analyzes the captured data using algorithms to map blood flow and oxygenation levels. == Clinical Application == The primary clinical application of this device is the rapid assessment of blood flow in the brain for: * '''Stroke Diagnosis''': Quick identification of ischemic regions where blood flow is compromised. * '''Monitoring Therapy''': Assessing the efficacy of interventions aimed at restoring normal blood flow. == Advantages Over Traditional Methods == The Openwater Blood Flow Analysis Device offers several benefits: * '''Non-invasive Operation''': Unlike invasive cerebral angiography, it poses no risks associated with surgical procedures. * '''Accessibility''': Its portability makes it suitable for bedside examinations and potential field use in emergency settings. * '''Real-time Analysis''': Provides immediate feedback on cerebral blood flow, crucial for urgent medical decision-making. * '''Speed of Testing''': Delivers diagnostic results in less significantly less time compared to traditional imaging methods like MRI, which can take much longer. == Safety Considerations == The device is designed with patient safety in mind and adheres to strict regulatory standards for medical devices. It includes fail-safes and alert systems to prevent any potential risks during operation. == Conclusion == The Openwater Blood Flow Analysis Device represents a significant innovation in medical diagnostics, offering a safe, efficient, and non-invasive method to evaluate cerebral blood flow. It has the potential to be a game-changer in the rapid detection and management of strokes and other conditions affecting cerebral hemodynamics. 6o3lnfm3hg77cilvpfbgi6mcsyjw9cc 625 294 2023-12-19T18:40:06Z Admin 1 Protected "[[Blood Flow Gen 2 Software]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 294 wikitext text/x-wiki = Openwater Gen2 Blood Flow Device Overview = == Introduction == The '''Openwater Blood Flow Analysis Device''' is an innovative diagnostic tool designed to non-invasively measure and monitor cerebral blood flow. Utilizing advanced near-infrared light techniques, it facilitates the detection of hemodynamic changes associated with various medical conditions, particularly cerebrovascular events like strokes. == Fundamental Theory == The device employs near-infrared spectroscopy (NIRS) to penetrate the skull and tissue, allowing for the visualization of blood flow within the brain. The key principles include: * '''Near-Infrared Light Penetration''': NIR light, due to its longer wavelength, can traverse the human skull and reach the brain, providing a window into cerebral circulation. * '''Hemodynamic Response Detection''': The device measures the dynamic changes in blood oxygenation and volume, which are indicative of the brain's metabolic activity. == Core Technologies == The device integrates several technological advancements to achieve precise blood flow imaging: * '''Diffuse Optical Tomography (DOT)''': Generates images by detecting light absorption and scattering patterns, which vary with blood flow and oxygenation. * '''Laser Speckle Contrast Imaging (LSCI)''': Measures blood flow velocity by analyzing the speckle pattern of reflected light from moving blood cells. == Device Components and Operation == The operational mechanism of the device is centered around a head mounted senor module that emits NIR light and collects returning signals: * '''Sensor Module''': A component that directs NIR light onto the patient's head and captures the backscattered light. * '''Processing Core''': Analyzes the captured data using algorithms to map blood flow and oxygenation levels. == Clinical Application == The primary clinical application of this device is the rapid assessment of blood flow in the brain for: * '''Stroke Diagnosis''': Quick identification of ischemic regions where blood flow is compromised. * '''Monitoring Therapy''': Assessing the efficacy of interventions aimed at restoring normal blood flow. == Advantages Over Traditional Methods == The Openwater Blood Flow Analysis Device offers several benefits: * '''Non-invasive Operation''': Unlike invasive cerebral angiography, it poses no risks associated with surgical procedures. * '''Accessibility''': Its portability makes it suitable for bedside examinations and potential field use in emergency settings. * '''Real-time Analysis''': Provides immediate feedback on cerebral blood flow, crucial for urgent medical decision-making. * '''Speed of Testing''': Delivers diagnostic results in less significantly less time compared to traditional imaging methods like MRI, which can take much longer. == Safety Considerations == The device is designed with patient safety in mind and adheres to strict regulatory standards for medical devices. It includes fail-safes and alert systems to prevent any potential risks during operation. == Conclusion == The Openwater Blood Flow Analysis Device represents a significant innovation in medical diagnostics, offering a safe, efficient, and non-invasive method to evaluate cerebral blood flow. It has the potential to be a game-changer in the rapid detection and management of strokes and other conditions affecting cerebral hemodynamics. 6o3lnfm3hg77cilvpfbgi6mcsyjw9cc 0 38 134 2023-12-13T03:22:45Z Gvigelet 4 Created page with " <span id="openwater-blood-flow-analysis-device---software-architecture-overview"></span> = Openwater Blood Flow Analysis Device - Software Architecture Overview = <span id="purpose"></span> == Purpose == This document provides an overview of the software architecture for the Openwater Blood Flow Analysis device, focusing on its key components and interactions. <span id="architecture-design"></span> == Architecture Design == The software architecture follows a layer..." 134 wikitext text/x-wiki <span id="openwater-blood-flow-analysis-device---software-architecture-overview"></span> = Openwater Blood Flow Analysis Device - Software Architecture Overview = <span id="purpose"></span> == Purpose == This document provides an overview of the software architecture for the Openwater Blood Flow Analysis device, focusing on its key components and interactions. <span id="architecture-design"></span> == Architecture Design == The software architecture follows a layered approach: * '''Presentation Layer''': User interface for interaction with the device. * '''Business Logic Layer''': Core functionality including data processing and analysis. * '''Data Access Layer''': Manages data storage and retrieval. <span id="main-components"></span> == Main Components == * '''User Interaction Module''': Handles user inputs and displays results. * '''Data Management Module''': Responsible for data storage and management. * '''Processing &amp; Analysis Module''': Processes blood flow data and generates analysis. * '''Device Control Module''': Manages device operations and settings. * '''Safety &amp; Alert Module''': Ensures device safety and alerts users of any issues. <span id="diagrams"></span> == Diagrams == High Level Software Component Diagram <p align="center"> [[File:ow_system_diagram.png|756x756px]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure: Openwater System Diagram|Openwater System Diagram</pre> </div> High Level Data Flow Diagram <p align="center"> [[File:ow_dataflow_diagram.png|781x781px]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure: Openwater Data Flow Diagram</pre> </div> <span id="use-cases"></span> == Use Cases == * Initiating blood flow analysis. * Viewing analysis results. * Adjusting device settings. * Responding to safety alerts. * Exporting data for external review. <span id="constraints"></span> == Constraints == * Hardware limitations. * Regulatory compliance. <span id="goals"></span> == Goals == * Reliability, usability, safety, scalability, interoperability, maintainability, security, and performance. h8hbnfx22fdgsp41jrja56ihcb1toci 626 134 2023-12-19T18:40:22Z Admin 1 Protected "[[Blood Flow Gen 2 Software Architecture]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 134 wikitext text/x-wiki <span id="openwater-blood-flow-analysis-device---software-architecture-overview"></span> = Openwater Blood Flow Analysis Device - Software Architecture Overview = <span id="purpose"></span> == Purpose == This document provides an overview of the software architecture for the Openwater Blood Flow Analysis device, focusing on its key components and interactions. <span id="architecture-design"></span> == Architecture Design == The software architecture follows a layered approach: * '''Presentation Layer''': User interface for interaction with the device. * '''Business Logic Layer''': Core functionality including data processing and analysis. * '''Data Access Layer''': Manages data storage and retrieval. <span id="main-components"></span> == Main Components == * '''User Interaction Module''': Handles user inputs and displays results. * '''Data Management Module''': Responsible for data storage and management. * '''Processing &amp; Analysis Module''': Processes blood flow data and generates analysis. * '''Device Control Module''': Manages device operations and settings. * '''Safety &amp; Alert Module''': Ensures device safety and alerts users of any issues. <span id="diagrams"></span> == Diagrams == High Level Software Component Diagram <p align="center"> [[File:ow_system_diagram.png|756x756px]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure: Openwater System Diagram|Openwater System Diagram</pre> </div> High Level Data Flow Diagram <p align="center"> [[File:ow_dataflow_diagram.png|781x781px]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure: Openwater Data Flow Diagram</pre> </div> <span id="use-cases"></span> == Use Cases == * Initiating blood flow analysis. * Viewing analysis results. * Adjusting device settings. * Responding to safety alerts. * Exporting data for external review. <span id="constraints"></span> == Constraints == * Hardware limitations. * Regulatory compliance. <span id="goals"></span> == Goals == * Reliability, usability, safety, scalability, interoperability, maintainability, security, and performance. h8hbnfx22fdgsp41jrja56ihcb1toci 0 17 51 2023-12-12T22:39:27Z Admin 1 Created page with "This is the camera test rig page!" 51 wikitext text/x-wiki This is the camera test rig page! 8fff023u4o1zrmaarjkzznvei8s8fqq 55 51 2023-12-12T22:57:07Z Admin 1 55 wikitext text/x-wiki This is the camera test rig page - to be updated 0cnwknf7kx0lkzx0umfcn5yl363zgni 187 55 2023-12-13T16:47:31Z 69.181.108.2 187 wikitext text/x-wiki This is the camera test rig page - to be updated ## Introduction The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. ## System Block Diagram The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair ### Camera tester BOM | qty | description | vendor | part number or spec | |-----|---------------------------|---------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | ... | ... | ... | ... | | ... | ... | ... | ... | ... ## Sequence of Operation 1. User powers up TDA4 and display, connects via serial or SSH. 2. Connects and powers on cameras. 3. Starts the program in command line mode and follows prompts for calibration. 4. Kills the program ('x' command) and disconnects cameras. 5. Return to step 2 for relevant cameras. 6. Powers down system. Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. ## Calibration Data During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: lh8rt5hv5qtk7tsllke1e2xwr58htp0 189 187 2023-12-13T16:58:31Z 69.181.108.2 189 wikitext text/x-wiki This is the camera test rig page - to be updated __TOC__ ## Introduction The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. ## System Block Diagram The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair ### Camera tester BOM | qty | description | vendor | part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | camera cover | OW | 3000-0803 | | 2 | microscope slide holders | SLH1 | thorlabs | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | --------------------------------------------------------------------------------- ... ## Sequence of Operation 1. User powers up TDA4 and display, connects via serial or SSH. 2. Connects and powers on cameras. 3. Starts the program in command line mode and follows prompts for calibration. 4. Kills the program ('x' command) and disconnects cameras. 5. Return to step 2 for relevant cameras. 6. Powers down system. Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. ## Calibration Data During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: orso2asj5fw0ylqygk84jctj3qd2i6j 190 189 2023-12-13T17:05:43Z Openwaterpete 5 190 wikitext text/x-wiki This is the camera test rig page - to be updated __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = --------------------------------------------------------------------------------- | qty | description | vendor | part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | camera cover | OW | 3000-0803 | | 2 | microscope slide holders | SLH1 | thorlabs | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | --------------------------------------------------------------------------------- ... = Sequence of Operationm= 1. User powers up TDA4 and display, connects via serial or SSH. 2. Connects and powers on cameras. 3. Starts the program in command line mode and follows prompts for calibration. 4. Kills the program ('x' command) and disconnects cameras. 5. Return to step 2 for relevant cameras. 6. Powers down system. Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: fba2k89ms4fzh16wrljprcqdsmrktfk 192 190 2023-12-13T17:07:55Z Openwaterpete 5 192 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = --------------------------------------------------------------------------------- | qty | description | vendor | part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | camera cover | OW | 3000-0803 | | 2 | microscope slide holders | SLH1 | thorlabs | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | --------------------------------------------------------------------------------- ... = Sequence of Operationm= 1. User powers up TDA4 and display, connects via serial or SSH. 2. Connects and powers on cameras. 3. Starts the program in command line mode and follows prompts for calibration. 4. Kills the program ('x' command) and disconnects cameras. 5. Return to step 2 for relevant cameras. 6. Powers down system. Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: hvh8p8aztkh9nk3fwn5y2yf7v0ejxa1 195 192 2023-12-13T17:10:31Z Openwaterpete 5 195 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = --------------------------------------------------------------------------------- | qty | description | vendor | part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | camera cover | OW | 3000-0803 | | 2 | microscope slide holders | SLH1 | thorlabs | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | --------------------------------------------------------------------------------- ... = Sequence of Operationm= <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: ccd84yygedvvht6bb1fjtea0kczhny7 199 195 2023-12-13T17:11:52Z Openwaterpete 5 199 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = --------------------------------------------------------------------------------- | qty | description | vendor | part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | camera cover | OW | 3000-0803 | | 2 | microscope slide holders | SLH1 | thorlabs | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | --------------------------------------------------------------------------------- ... = Sequence of Operationm= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: r86od0m0l1rgqb601hw2a0vpw0vinpo 200 199 2023-12-13T17:15:01Z Openwaterpete 5 200 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = | Qty | Description | Vendor | Part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | Camera cover | OW | 3000-0803 | | 2 | Microscope slide holders | Thorlabs | SLH1 | | 1 | | | | | 1 | | | | | Qty | Description | Vendor | Part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | camera cover | OW | 3000-0803 | | 2 | microscope slide holders | SLH1 | thorlabs | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | --------------------------------------------------------------------------------- ... = Sequence of Operationm= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: k7gnr06glew6l83mndrc5xs2fihr0pf 201 200 2023-12-13T17:17:37Z Openwaterpete 5 201 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = {| class="wikitable" |- | Qty | Description | Vendor | Part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | Camera cover | OW | 3000-0803 | | 2 | Microscope slide holders | Thorlabs | SLH1 | | 1 | | | | | 1 | | | | | Qty | Description | Vendor | Part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | camera cover | OW | 3000-0803 | | 2 | microscope slide holders | SLH1 | thorlabs | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | --------------------------------------------------------------------------------- ... = Sequence of Operationm= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: c6pdm044a51xpul86ejlneboigxvdje 202 201 2023-12-13T17:19:19Z Openwaterpete 5 202 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = {| class="wikitable" |- | Qty | Description | Vendor | Part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | camera cover | OW | 3000-0803 | | 2 | microscope slide holders | SLH1 | thorlabs | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | --------------------------------------------------------------------------------- |} ... = Sequence of Operationm= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: 6pgipnzkl7k6mqzig3038go8boe0k5i 203 202 2023-12-13T17:19:46Z Openwaterpete 5 /* Sequence of Operationm */ 203 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = {| class="wikitable" |- | Qty | Description | Vendor | Part number or spec | |-----|---------------------------|-----------------------|---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | camera cover | OW | 3000-0803 | | 2 | microscope slide holders | SLH1 | thorlabs | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | --------------------------------------------------------------------------------- |} ... = Sequence of Operation= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: 4nf8el2wxvyo7t12mz9jtukirshbfmu 206 203 2023-12-13T17:23:28Z Openwaterpete 5 206 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = {| class="wikitable" |- | Qty || Description || Vendor || Part number or spec | |- | |- |-----||---------------------------||-----------------------||---------------------| | 1 | Processor (TDA4) | D3 | 7000-0162 | | 1 | Uniform illuminator | Advanced Illumination | BL2-0404 whiiS3 | | 1 | camera cover | OW | 3000-0803 | | 2 | microscope slide holders | SLH1 | thorlabs | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | | 1 | | | | --------------------------------------------------------------------------------- |} ... = Sequence of Operation= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: 3c0cchae0qvq5609dsrf276rd8iniib 208 206 2023-12-13T17:26:32Z Openwaterpete 5 208 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = {| class="wikitable" |- | Qty || Description || Vendor || Part number or spec || |- | |- |-----||---------------------------||-----------------------||---------------------|| | 1 || Processor (TDA4) || D3 || 7000-0162 || | 1 || Uniform illuminator || Advanced Illumination || BL2-0404 whiiS3 || | 1 || camera cover || OW ||3000-0803 || | 2 || microscope slide holders || SLH1 || thorlabs || | 1 || || || || | 1 || || || || | 1 || || || || | 1 || || || || | 1 || || || || | 1 || || || || --------------------------------------------------------------------------------- |} ... = Sequence of Operation= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: may5f08q4mo5k390p6gvb791gfwxbnr 209 208 2023-12-13T17:27:35Z Openwaterpete 5 209 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = {| class="wikitable" |- | Qty || Description || Vendor || Part number or spec || |- | |- |-----||---------------------------||-----------------------||---------------------|| | 1 || Processor (TDA4) || D3 || 7000-0162 || |- | 1 || Uniform illuminator || Advanced Illumination || BL2-0404 whiiS3 || |- | 1 || camera cover || OW ||3000-0803 || |- | 2 || microscope slide holders || SLH1 || thorlabs || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- --------------------------------------------------------------------------------- |} ... = Sequence of Operation= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: gus7p31a9fh2mtr8mhos7kesx11tfek 210 209 2023-12-13T17:29:19Z Openwaterpete 5 210 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = {| class="wikitable" |- | Qty || Description || Vendor || Part number or spec || |- | |- |-----||---------------------------||-----------------------||---------------------|| | 1 || Processor (TDA4) || D3 || 7000-0162 || |- | 1 || Uniform illuminator || Advanced Illumination || BL2-0404 whiiS3 || |- | 1 || camera cover || OW ||3000-0803 || |- | 2 || microscope slide holders || SLH1 || thorlabs || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- --------------------------------------------------------------------------------- |} ... = Sequence of Operation= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: 4y64d29d7dvem0rd5t5yabe0k7jht6y 214 210 2023-12-13T18:00:45Z 50.227.118.138 214 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ <math> f(x) = x^2\,</math> = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = {| class="wikitable" |- | Qty || Description || Vendor || Part number or spec || |- | |- |-----||---------------------------||-----------------------||---------------------|| | 1 || Processor (TDA4) || D3 || 7000-0162 || |- | 1 || Uniform illuminator || Advanced Illumination || BL2-0404 whiiS3 || |- | 1 || camera cover || OW ||3000-0803 || |- | 2 || microscope slide holders || SLH1 || thorlabs || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- --------------------------------------------------------------------------------- |} ... = Sequence of Operation= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: a3v7vk2y1ir395iifx7jrez090k77vv 223 214 2023-12-13T19:12:03Z OpenwaterEthan 7 223 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ = Introduction = The camera test rig runs a branch of the device firmware from mid-2022... The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates relevant quantities about the connected cameras for calibration purposes. The user initiates the program in command line mode to gather calibration data from numerous cameras quickly, testing compliance after manufacturing. =System Block Diagram= The system consists of: - TDA4 processor - Up to 8 cameras connected via FPD Link - Uniform light source - 3D printed jig to hold cameras - Actuated TDA4 via SSH-enabled computer - Optional: Barcode scanner to label calibration outputs - Display connected to TDA4's Mini DisplayPort output for viewing selected camera pair = Camera tester BOM = {| class="wikitable" |- | Qty || Description || Vendor || Part number or spec || |- | |- |-----||---------------------------||-----------------------||---------------------|| | 1 || Processor (TDA4) || D3 || 7000-0162 || |- | 1 || Uniform illuminator || Advanced Illumination || BL2-0404 whiiS3 || |- | 1 || camera cover || OW ||3000-0803 || |- | 2 || microscope slide holders || SLH1 || thorlabs || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- | 1 || || || || |- --------------------------------------------------------------------------------- |} ... = Sequence of Operation= <ol> <li>User powers up TDA4 and display, connects via serial or SSH.</li> <li>Connects and powers on cameras.</li> <li>Starts the program in command line mode and follows prompts for calibration.</li> <li>Kills the program ('x' command) and disconnects cameras.</li> <li>Return to step 2 for relevant cameras.</li> <li>Powers down system.</li> </ol> Note: Different cameras may have different relevant gain settings; two precompiled versions of the system are available for convenience. Camera hot swapping is not supported. = Calibration Data = During calibration, three images are taken: two dark, one light. Each image is 2100x1450 with grayscale values between 0 and 1023. The first 20 rows of the sensor are painted over for calibration and remain dark. Quantities calculated from these images include: - Dark Pixels - Dead Pixels - Dark Image Mean / Variance - Light Image Mean / Variance - Read Noise - Device Temperature (uncalibrated) Example JSON output of a camera calibration is included in the repo. An image from the rig may look like this: 9z5v8r74uynvhnt59ty1pnnxxi13erd 253 223 2023-12-13T20:30:32Z OpenwaterEthan 7 Added contents from Google Doc 253 wikitext text/x-wiki This is the camera test rig page - to be updated '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' Camera Test Rig The camera test rig runs a branch of the device firmware from mid 2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. == System Block Diagram == The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig may look like this: __TOC__ h7rbmtqn5ey8fux3ehkteef8fiefjxm 254 253 2023-12-13T20:31:17Z OpenwaterEthan 7 254 wikitext text/x-wiki The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' == System Block Diagram == The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig may look like this: __TOC__ rqmam6lkb3i99r8sv446y9j4ndwfqau 269 254 2023-12-13T21:37:03Z OpenwaterEthan 7 269 wikitext text/x-wiki [[File:Camera test rig.png]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' == System Block Diagram == The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig may look like this: __TOC__ 2l576qd9cncy7iqlxwq7ntnvsho27qt 270 269 2023-12-13T21:37:36Z OpenwaterEthan 7 270 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' == System Block Diagram == The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig may look like this: __TOC__ ll70otfvwi3nbg81m6qz09xw6165vg0 271 270 2023-12-13T21:38:05Z OpenwaterEthan 7 271 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right| Camera test setup fully configured in the lab]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' == System Block Diagram == The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig may look like this: __TOC__ qvv7by2kvrkidcct7vclcusby1cxdqz 272 271 2023-12-13T21:38:38Z OpenwaterEthan 7 272 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right| Camera test setup fully configured in the lab]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ == System Block Diagram == The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig may look like this: mzcpw6ntljqa47iyt4ta2cevwwx1zbg 276 272 2023-12-13T21:41:01Z OpenwaterEthan 7 276 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right| Camera test setup fully configured in the lab]] [[File:Camera Holder CAD.png|thumb|Screenshot of the CAD of the camera holder]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ == System Block Diagram == [[File:Camera test rig block diagram.png|thumb]] The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig may look like this: h3vqfj46hoay4ll28toabg7gebq1rrd 278 276 2023-12-13T21:41:23Z OpenwaterEthan 7 278 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right| Camera test setup fully configured in the lab]] [[File:Camera Holder CAD.png|thumb|Screenshot of the CAD of the camera holder]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ == System Block Diagram == [[File:Camera test rig block diagram.png]] The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig may look like this: 534n2tj6k1g19xse5a4swlqkdyhz33b 280 278 2023-12-13T21:41:39Z OpenwaterEthan 7 280 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right| Camera test setup fully configured in the lab]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ == System Block Diagram == [[File:Camera test rig block diagram.png]] The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == [[File:Camera Holder CAD.png|thumb|Screenshot of the CAD of the camera holder]] {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig may look like this: 6a1xf9jgl1fcyjr06yt4v01fk4gxbam 285 280 2023-12-13T21:46:24Z OpenwaterEthan 7 285 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right| Camera test setup fully configured in the lab]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ == System Block Diagram == [[File:Camera test rig block diagram.png]] The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == [[File:Camera Holder CAD.png|thumb|Screenshot of the CAD of the camera holder]] {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == [[File:Test frame.png|thumb|600x600px|Image from the camera test rig of a "far" style camera with the illumination source on]] During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig can be seen on the right. 8tlsz4t19db56wlhte5tlo69zbdotal 286 285 2023-12-13T21:46:57Z OpenwaterEthan 7 286 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right| Camera test setup fully configured in the lab]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ == System Block Diagram == [[File:Camera_test_rig_block_diagram.png]] The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == [[File:Camera Holder CAD.png|thumb|Screenshot of the CAD of the camera holder]] {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == [[File:Test frame.png|thumb|600x600px|Image from the camera test rig of a "far" style camera with the illumination source on]] During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig can be seen on the right. fp6omhfyoa7knr7dosvqgsm9y7ynerk 287 286 2023-12-13T21:47:19Z OpenwaterEthan 7 287 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right| Camera test setup fully configured in the lab]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ == System Block Diagram == [[File:Camera_test_rig_block_diagram.png]] The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == [[File:Camera Holder CAD.png|thumb|Screenshot of the CAD of the camera holder]] {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == [[File:Test frame.png|thumb|600x600px|Image from the camera test rig of a "far" style camera with the illumination source on]] During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig can be seen on the right. oia29y03e6upf152yo34yg1gk9e7b8s 288 287 2023-12-13T21:51:06Z OpenwaterEthan 7 288 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right| Camera test setup fully configured in the lab]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ == System Block Diagram == [[File:Camera_test_rig_block_diagram.png|border|frameless|738x738px]] The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == [[File:Camera Holder CAD.png|thumb|Screenshot of the CAD of the camera holder]] {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == [[File:Test frame.png|thumb|600x600px|Image from the camera test rig of a "far" style camera with the illumination source on]] During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig can be seen on the right. f1hbzrl0b37898q9kll3lpxzg8sjp24 627 288 2023-12-19T18:40:36Z Admin 1 Protected "[[Camera Test Rig]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 288 wikitext text/x-wiki [[File:Camera test rig.png|thumb|right| Camera test setup fully configured in the lab]] The camera test rig runs a branch of the device firmware from mid-2022 that streams camera imagery to a connected display, captures images, and calculates a number of relevant quantities about the connected cameras for calibration purposes. The user is meant to start up the program in command line mode and use the system to gather calibration data from a large number of cameras quickly after a round of manufacturing to test compliance. '''Please visit the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_camera_tester/tree/main GitHub repository] repository for the complete collection of files referenced on this page.''' __TOC__ == System Block Diagram == [[File:Camera_test_rig_block_diagram.png|border|frameless|738x738px]] The system consists of the TDA4, up to 8 cameras connected via FPD Link, and uniform light source. The cameras are held in place using a 3D printed jig and the TDA4 is actuated via any computer that can SSH in. Optionally, for ease of use, a barcode scanner can be connected to the host computer to scan in the serial numbers of the included cameras for labeling the calibration outputs. A display can also be connected to the TDA4's Mini DisplayPort output for viewing the selected camera pair. == Camera tester BOM == [[File:Camera Holder CAD.png|thumb|Screenshot of the CAD of the camera holder]] {| class="wikitable" |'''Qty''' |'''Description''' |'''Vendor''' |'''Part Number or spec''' |- |1 |processor |D3 |7000-0162 |- |1 |uniform illuminator |advanced illumination |BL2-0404 whiiS3 |- |1 |camera cover |OW |3000-0803 |- |2 |microscope slide holders |thorlabs |SLH1 |- |1 |lighting- dual source |keysight |24V and 1-10 V |- |1 |aggregator power supply |phihong |5V, 1.5A |- |1 |display camera |FDI |eli 101 |- |1 |display TDA4 |samsung | |- |1 |aggregator board |OW |7000-0136 |- |1 |PC |Dell | |- |1 |keyboard/mouse |Dell | |- |1 |barcode scanner |tera scanner |D5100y |- |4 |camera power cable |CSE |3000-0548 |- |1 |display for PC |dell | |- |4 |camera HS coax cable |pasternack |PE39350Z |- |4 |camera coax |CSE |7000-0158 |- |1 |ethernet cable |digikey | |- |1 |cable enclosure box |mcmaster |7593K32 |- |1 |veil |buffalo company | |} == Sequence of Operation == 1. User powers up TDA4 and display and connects to the system via serial or SSH 2. User connects cameras to TDA4 and powers them on 3. User starts the program in command line mode and follows the prompts to start the calibration process. 4. User kills the program (using the 'x' command) and disconnects the cameras. 5. Return to step 2 for as many cameras as are relevant. 6. User powers down system Note: different cameras have different relevant gain settings so for convenience and quick software development, two precompiled versions of the system are available. '''Camera hot swapping at this time is not supported.''' == Calibration Data == [[File:Test frame.png|thumb|600x600px|Image from the camera test rig of a "far" style camera with the illumination source on]] During calibration, three images are taken: two with the illumination source off (dark) and one with it on (light). Each of the images is 2100x1450 and has a grayscale value between 0 and 1023. The first 20 rows of the image sensor are painted over for calibration purposes and will always remain dark. The average pixel value in this region and the "dark" image should be about the same and may increase with camera temperature. The following quantities are calculated from these three images: '''Dark Pixels''' - count of number of pixels with a value of 0 '''Dead Pixels''' - count of number of pixels whose value do not change by 10 or less between the dark image and the light image. These can be cases where there is dust from manufacturing on the image sensor or where the pixel is just "stuck" '''Dark Image Mean / Variance''' - across one of the dark frames, calculate the mean pixel value and the variance in pixel values '''Light image mean / variance''' - across one of the light frames, calculate the mean pixel value and the variance in pixel values '''Read noise''' - calculates the difference in variance from the two captured dark frames, cuts it in half and takes the square root to get a metric for how much the dark pixels change between captures '''100px histo width''' - counts the number of bins with count greater than 100 '''Device temperature''' - temperature of the camera sensor as read from registers. Note that this is uncalibrated and there is nontrivial variation between sensors. An example of the JSON output of a camera calibration is included in the repo. An image from the rig can be seen on the right. f1hbzrl0b37898q9kll3lpxzg8sjp24 0 122 616 2023-12-19T18:16:52Z Admin 1 Created page with "Community projects may use this space to share documentation and information about their work. Users are encouraged to link to their GitHub repositories, documentation, and pages of their own creation on this Wiki. This page is editable to all users." 616 wikitext text/x-wiki Community projects may use this space to share documentation and information about their work. Users are encouraged to link to their GitHub repositories, documentation, and pages of their own creation on this Wiki. This page is editable to all users. b6x1muxwx0le266nnhg1ast57d00yki 617 616 2023-12-19T18:17:56Z Admin 1 617 wikitext text/x-wiki Community projects may use this space to share documentation and information about their work. Users are encouraged to link to their GitHub repositories, documentation, and pages of their own creation on this Wiki. This page is editable to all users. == Registry == No open source projects / forks to link to (yet!) tpvyl1fybb5f1pxu876ol1a2qwg4fk5 0 19 56 2023-12-12T23:09:12Z Opw12 8 Acousto Optic 56 wikitext text/x-wiki <span id="Background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="Technology Overview"></span> = Technology-Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the hardware and software repositories on GitHub. 50irhdrxmrzj0rtmfr6z4tsgez1ibha 58 56 2023-12-12T23:20:38Z Opw12 8 58 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the hardware and software repositories on GitHub. <p align="center"> [[File:opw_acousto-optic - High level diagram.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Text here.</pre> </div> </p> <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic data repository on GitHub. <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Embedded Vein Phantom with Coregistered MRI (795 nm CW laser light, 5 MHz fixed focus transducer).</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer) .</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer).</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Ex Vivo Rat Head (795 nm CW laser light, 5 MHz fixed focus transducer).</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: In Vivo Rat with Large Subcutaneous Allogeneic Tumor (830 nm laser illumination with 150 ns pulse length, 5 MHz fixed focus transducer).</pre> </div> </p> <span id="references"></span> == References == fz2wcrtd04ykxoq7k19x0p46ppz9o61 62 58 2023-12-12T23:30:02Z Opw12 8 62 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the hardware and software repositories on GitHub. <p align="center"> [[File:opw_acousto-optic_-_High_level_diagram.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Text here.</pre> </div> </p> <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic data repository on GitHub. <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Embedded Vein Phantom with Coregistered MRI (795 nm CW laser light, 5 MHz fixed focus transducer).</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer) .</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer).</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Ex Vivo Rat Head (795 nm CW laser light, 5 MHz fixed focus transducer).</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: In Vivo Rat with Large Subcutaneous Allogeneic Tumor (830 nm laser illumination with 150 ns pulse length, 5 MHz fixed focus transducer).</pre> </div> </p> <span id="references"></span> == References == abbootyihce6cknm4j0mbyreird7rfz 63 62 2023-12-12T23:31:36Z Opw12 8 /* Technology Overview */ 63 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the hardware and software repositories on GitHub. <p align="center"> [[File:opw_acousto-optic_-_High_level_diagram.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Text here.</pre> </div> </p> <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic data repository on GitHub. [[File:Ex Vivo Rat Torso.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Embedded Vein Phantom with Coregistered MRI (795 nm CW laser light, 5 MHz fixed focus transducer).</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer) .</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer).</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: Ex Vivo Rat Head (795 nm CW laser light, 5 MHz fixed focus transducer).</pre> </div> </p> <p align="center"> [[File:figureX.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure X: In Vivo Rat with Large Subcutaneous Allogeneic Tumor (830 nm laser illumination with 150 ns pulse length, 5 MHz fixed focus transducer).</pre> </div> </p> <span id="references"></span> == References == cnjugex1lx0zx0fmgvmfi8knuu7tjw2 64 63 2023-12-12T23:37:50Z Opw12 8 64 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the hardware and software repositories on GitHub. [[File:opw_acousto-optic_-_High_level_diagram.png|center|frame|Acousto-Optic High Level Diagram]] <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic data repository on GitHub. [[File:Ex Vivo Rat Torso.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] <span id="references"></span> == References == bx9f9jvrybkpbmiqnbajcu4otkrdjft 68 64 2023-12-12T23:41:26Z Opw12 8 /* Results from this System */ 68 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the hardware and software repositories on GitHub. [[File:opw_acousto-optic_-_High_level_diagram.png|center|frame|Acousto-Optic High Level Diagram]] <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic data repository on GitHub. [[File:Ex Vivo Rat Torso.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Porcine Kidneys.gif|center|frame|Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Head.png|center|frame|Ex Vivo Rat Head]] [[File:In Vivo Rat with Large Subcutaneous Tumor.gif|center|frame|In Vivo Rat with Large Subcutaneous Tumor]] <span id="references"></span> == References == j031jccvedx6hef0uh5ir6h726rpj8n 69 68 2023-12-12T23:43:54Z Opw12 8 69 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the [https://github.com/OpenwaterInternet/opw_acousto-optic_hw/tree/main hardware] and [https://github.com/OpenwaterInternet/opw_acousto-optic_sw/tree/main software] repositories on GitHub. [[File:opw_acousto-optic_-_High_level_diagram.png|center|frame|Acousto-Optic High Level Diagram]] <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic [https://github.com/OpenwaterInternet/opw_acousto-optic_data/tree/main data] repository on GitHub. [[File:Ex Vivo Rat Torso.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Porcine Kidneys.gif|center|frame|Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Head.png|center|frame|Ex Vivo Rat Head]] [[File:In Vivo Rat with Large Subcutaneous Tumor.gif|center|frame|In Vivo Rat with Large Subcutaneous Tumor]] <span id="references"></span> == References == 5dwz8kl852kc1r7m60rzv7xmin6ia06 73 69 2023-12-12T23:51:05Z Opw12 8 73 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the [https://github.com/OpenwaterInternet/opw_acousto-optic_hw/tree/main hardware] and [https://github.com/OpenwaterInternet/opw_acousto-optic_sw/tree/main software] repositories on GitHub. [[File:opw_acousto-optic_-_High_level_diagram.png|center|frame|Acousto-Optic High Level Diagram]] <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic [https://github.com/OpenwaterInternet/opw_acousto-optic_data/tree/main data] repository on GitHub. [[File:Ex_Vivo_Rat_Torso.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Torso - better.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Porcine Kidneys.gif|center|frame|Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Head.png|center|frame|Ex Vivo Rat Head]] [[File:In Vivo Rat with Large Subcutaneous Tumor.gif|center|frame|In Vivo Rat with Large Subcutaneous Tumor]] <span id="references"></span> == References == bz3ciqvxbiet3hb03fxu8o9xk778msc 75 73 2023-12-12T23:52:48Z Opw12 8 75 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the [https://github.com/OpenwaterInternet/opw_acousto-optic_hw/tree/main hardware] and [https://github.com/OpenwaterInternet/opw_acousto-optic_sw/tree/main software] repositories on GitHub. [[File:opw_acousto-optic_-_High_level_diagram.png|center|frame|Acousto-Optic High Level Diagram]] <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic [https://github.com/OpenwaterInternet/opw_acousto-optic_data/tree/main data] repository on GitHub. [[File:Ex_Vivo_Rat_Torso.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Torso - better.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Porcine Kidneys.gif|center|frame|Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Head.png|center|frame|Ex Vivo Rat Head]] [[File:In Vivo Rat with Large Subcutaneous Tumor.gif|center|frame|In Vivo Rat with Large Subcutaneous Tumor]] <span id="references"></span> == References == 6zjy5wl5uocf7i19c9tn90s9jo9h6os 84 75 2023-12-13T00:11:20Z Opw12 8 84 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the [https://github.com/OpenwaterInternet/opw_acousto-optic_hw/tree/main hardware] and [https://github.com/OpenwaterInternet/opw_acousto-optic_sw/tree/main software] repositories on GitHub. [[File:opw_acousto-optic_-_High_level_diagram.png|center|frame|Acousto-Optic High Level Diagram]] <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic [https://github.com/OpenwaterInternet/opw_acousto-optic_data/tree/main data] repository on GitHub. [[File:Embedded Vein Phantom - OW small.gif|center|frame|Embedded Vein Phantom with Coregistered MRI (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Embedded Vein Phantom - MRI small.gif|center|frame|Embedded Vein Phantom with Coregistered MRI (MRI scanner)]] [[File:Ex_Vivo_Rat_Torso.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Torso - better.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Porcine Kidneys.gif|center|frame|Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Head.png|center|frame|Ex Vivo Rat Head]] [[File:In Vivo Rat with Large Subcutaneous Tumor.gif|center|frame|In Vivo Rat with Large Subcutaneous Tumor]] <span id="references"></span> == References == 147nxn8hjcws0r7nctrc6fmnhazz79e 85 84 2023-12-13T00:16:10Z Opw12 8 85 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the [https://github.com/OpenwaterInternet/opw_acousto-optic_hw/tree/main hardware] and [https://github.com/OpenwaterInternet/opw_acousto-optic_sw/tree/main software] repositories on GitHub. [[File:opw_acousto-optic_-_High_level_diagram.png|center|frame|Acousto-Optic High Level Diagram]] <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic [https://github.com/OpenwaterInternet/opw_acousto-optic_data/tree/main data] repository on GitHub. [[File:Embedded Vein Phantom - OW small.gif|center|frame|Embedded Vein Phantom (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Embedded Vein Phantom - MRI small.gif|center|frame|Embedded Vein Phantom (MRI scanner)]] [[File:Ex_Vivo_Rat_Torso.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Torso - better.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Porcine Kidneys.gif|center|frame|Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Head.png|center|frame|Ex Vivo Rat Head (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:In Vivo Rat with Large Subcutaneous Tumor.gif|center|frame|In Vivo Rat with Large Subcutaneous Allogeneic Tumor (830 nm laser illumination with 150 ns pulse length, 5 MHz fixed focus transducer)]] <span id="references"></span> == References == l99dhumob4czrw1bc4d7ba01l5l32s3 337 85 2023-12-14T00:26:45Z Opw12 8 Opw12 moved page [[Acousto Optic]] to [[Holographic Acousto Optic Imaging]] 85 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the [https://github.com/OpenwaterInternet/opw_acousto-optic_hw/tree/main hardware] and [https://github.com/OpenwaterInternet/opw_acousto-optic_sw/tree/main software] repositories on GitHub. [[File:opw_acousto-optic_-_High_level_diagram.png|center|frame|Acousto-Optic High Level Diagram]] <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic [https://github.com/OpenwaterInternet/opw_acousto-optic_data/tree/main data] repository on GitHub. [[File:Embedded Vein Phantom - OW small.gif|center|frame|Embedded Vein Phantom (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Embedded Vein Phantom - MRI small.gif|center|frame|Embedded Vein Phantom (MRI scanner)]] [[File:Ex_Vivo_Rat_Torso.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Torso - better.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Porcine Kidneys.gif|center|frame|Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Head.png|center|frame|Ex Vivo Rat Head (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:In Vivo Rat with Large Subcutaneous Tumor.gif|center|frame|In Vivo Rat with Large Subcutaneous Allogeneic Tumor (830 nm laser illumination with 150 ns pulse length, 5 MHz fixed focus transducer)]] <span id="references"></span> == References == l99dhumob4czrw1bc4d7ba01l5l32s3 620 337 2023-12-19T18:38:16Z Admin 1 Protected "[[Holographic Acousto Optic Imaging]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 85 wikitext text/x-wiki <span id="background"></span> = Background = One of the most common challenges across much of medical imaging is the ability to capture both structural information, e.g. the shapes and locations of musculoskeletal tissues, as well as functional information, e.g. how much blood is perfusing through these very same tissues. Near infrared spectroscopy (NIRS) is a well established field of medical imaging that uses light to provide a wide variety of functional information (pulse-oximetry is an example of this type of measurement). NIRS however suffers from very poor spatial resolution due to the highly scattering nature of light with tissue. Conversely, ultrasound is a well established method that can provide great structural information of the same tissues, albeit providing very limited functional information. Holographic acousto-optic imaging is a unique hybrid method to combine these two modalities in order to achieve functional information at high spatial resolution. <span id="technology-overview"></span> = Technology Overview = Holographic acousto-optic imaging works by first directing light from a laser beam into tissue. When this light travels inside the tissue, it will be rapidly diffusely scattered and ultimately make it challenging to obtain any structural information from it. However, when an ultrasound beam’s focus is placed within the tissue at the same time, the laser light that crosses that focus region will be shifted to a very slightly different wavelength, this essentially ‘tags’ the light. By using lasers with very specific properties (notably high coherence), one can sort out this ‘tagged’ light from all of the other light exiting the tissue by holographically recombining it with some of the original laser light, similarly shifted in wavelength. Thus, holographic acousto-optic imaging can provide useful functional information at great spatial detail. This has been diagrammed in the image below. By moving the ultrasound focus across the entire tissue area, one can generate a 3D image of the functional properties of the tissue. Greater detail on the Openwater holographic acousto-optic image setup can be found in the [https://github.com/OpenwaterInternet/opw_acousto-optic_hw/tree/main hardware] and [https://github.com/OpenwaterInternet/opw_acousto-optic_sw/tree/main software] repositories on GitHub. [[File:opw_acousto-optic_-_High_level_diagram.png|center|frame|Acousto-Optic High Level Diagram]] <span id="results-from-this-system"></span> = Results from this System = Below is a listing of select imaging scans acquired with Openwater acousto-optic setups. Many of these along with other datasets can be found on the acousto-optic [https://github.com/OpenwaterInternet/opw_acousto-optic_data/tree/main data] repository on GitHub. [[File:Embedded Vein Phantom - OW small.gif|center|frame|Embedded Vein Phantom (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Embedded Vein Phantom - MRI small.gif|center|frame|Embedded Vein Phantom (MRI scanner)]] [[File:Ex_Vivo_Rat_Torso.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Torso - better.gif|center|frame|Ex Vivo Rat Torso (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Porcine Kidneys.gif|center|frame|Ex Vivo Porcine Kidneys (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:Ex Vivo Rat Head.png|center|frame|Ex Vivo Rat Head (795 nm CW laser light, 5 MHz fixed focus transducer)]] [[File:In Vivo Rat with Large Subcutaneous Tumor.gif|center|frame|In Vivo Rat with Large Subcutaneous Allogeneic Tumor (830 nm laser illumination with 150 ns pulse length, 5 MHz fixed focus transducer)]] <span id="references"></span> == References == l99dhumob4czrw1bc4d7ba01l5l32s3 0 26 72 2023-12-12T23:49:45Z Openwaterpete 5 Created page with "'''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure they meet performance requirements. Long coherence length (narrow linewidth) is critical. Refer to Gen 1 Blood Flow White Paper for detaile..." 72 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure they meet performance requirements. Long coherence length (narrow linewidth) is critical. Refer to Gen 1 Blood Flow White Paper for detailed background. == Background == Producing high-power illumination with the required linewidth for blood flow detection is challenging. Operating in pulsed mode, as opposed to continuous wave (CW), complicates maintaining stable linewidths. == Coherence Length == The coherence length measures the spatial extent where laser electromagnetic waves maintain consistent phase relationships. It determines interference effects and our need for well-defined patterns in blood flow detection. == Chirp == Laser chirp refers to frequency or wavelength variations over time. Managing chirp is vital for stable speckle signals in our blood flow measurement. == Interferometer Experiment == We used interferometric techniques to compare coherence, frequency, and power stability. The experiment details and sample data are in the repository. == Speckle Contrast Measurements == A direct speckle contrast measurement test assesses coherence length's effect on speckle contrast. High coherence (narrower linewidth) improves blood flow detection. === Effect of Chirping on Speckle Contrast === Figure 2 demonstrates the detrimental effect of chirping on speckle contrast, necessitating stable laser frequency for accurate measurements. == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” g5feadwjl075dwo8jbobrcborxzziyh 74 72 2023-12-12T23:51:58Z Openwaterpete 5 /* Introduction */ 74 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure they meet performance requirements. Long coherence length (narrow linewidth) is critical. Refer to Gen 1 Blood Flow White Paper for detailed background. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely a) Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. b) Speckle contrast measurements. == Background == Producing high-power illumination with the required linewidth for blood flow detection is challenging. Operating in pulsed mode, as opposed to continuous wave (CW), complicates maintaining stable linewidths. == Coherence Length == The coherence length measures the spatial extent where laser electromagnetic waves maintain consistent phase relationships. It determines interference effects and our need for well-defined patterns in blood flow detection. == Chirp == Laser chirp refers to frequency or wavelength variations over time. Managing chirp is vital for stable speckle signals in our blood flow measurement. == Interferometer Experiment == We used interferometric techniques to compare coherence, frequency, and power stability. The experiment details and sample data are in the repository. == Speckle Contrast Measurements == A direct speckle contrast measurement test assesses coherence length's effect on speckle contrast. High coherence (narrower linewidth) improves blood flow detection. === Effect of Chirping on Speckle Contrast === Figure 2 demonstrates the detrimental effect of chirping on speckle contrast, necessitating stable laser frequency for accurate measurements. == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 5xqe6zkio77ho8c96exc1euzy9wiiod 89 74 2023-12-13T00:40:16Z Openwaterpete 5 89 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure they meet performance requirements. Long coherence length (narrow linewidth) is critical. Refer to Gen 1 Blood Flow White Paper for detailed background. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely a) Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. b) Speckle contrast measurements. == Background == Producing high-power illumination with the required linewidth for blood flow detection is challenging. Operating in pulsed mode, as opposed to continuous wave (CW), complicates maintaining stable linewidths. == Coherence Length == The coherence length measures the spatial extent where laser electromagnetic waves maintain consistent phase relationships. It determines interference effects and our need for well-defined patterns in blood flow detection. == Chirp == Laser chirp refers to frequency or wavelength variations over time. Managing chirp is vital for stable speckle signals in our blood flow measurement. == Interferometer Experiment == We used interferometric techniques to compare coherence, frequency, and power stability. The experiment details and sample data are in the repository. == Speckle Contrast Measurements == A direct speckle contrast measurement test assesses coherence length's effect on speckle contrast. High coherence (narrower linewidth) improves blood flow detection. == Effect of Chirping on Speckle Contrast == Figure 2 demonstrates the detrimental effect of chirping on speckle contrast, necessitating stable laser frequency for accurate measurements. == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” ac4aloagdylsgivl9afssj958aep40s 160 89 2023-12-13T13:58:19Z OpenwaterAndrew 3 /* Introduction */ 160 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == Producing high-power illumination with the required linewidth for blood flow detection is challenging. Operating in pulsed mode, as opposed to continuous wave (CW), complicates maintaining stable linewidths. == Coherence Length == The coherence length measures the spatial extent where laser electromagnetic waves maintain consistent phase relationships. It determines interference effects and our need for well-defined patterns in blood flow detection. == Chirp == Laser chirp refers to frequency or wavelength variations over time. Managing chirp is vital for stable speckle signals in our blood flow measurement. == Interferometer Experiment == We used interferometric techniques to compare coherence, frequency, and power stability. The experiment details and sample data are in the repository. == Speckle Contrast Measurements == A direct speckle contrast measurement test assesses coherence length's effect on speckle contrast. High coherence (narrower linewidth) improves blood flow detection. == Effect of Chirping on Speckle Contrast == Figure 2 demonstrates the detrimental effect of chirping on speckle contrast, necessitating stable laser frequency for accurate measurements. == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” ndvuxv8ifkcgkoo18jw0e57lb9nzyeo 161 160 2023-12-13T13:58:37Z OpenwaterAndrew 3 /* Background */ 161 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length measures the spatial extent where laser electromagnetic waves maintain consistent phase relationships. It determines interference effects and our need for well-defined patterns in blood flow detection. == Chirp == Laser chirp refers to frequency or wavelength variations over time. Managing chirp is vital for stable speckle signals in our blood flow measurement. == Interferometer Experiment == We used interferometric techniques to compare coherence, frequency, and power stability. The experiment details and sample data are in the repository. == Speckle Contrast Measurements == A direct speckle contrast measurement test assesses coherence length's effect on speckle contrast. High coherence (narrower linewidth) improves blood flow detection. == Effect of Chirping on Speckle Contrast == Figure 2 demonstrates the detrimental effect of chirping on speckle contrast, necessitating stable laser frequency for accurate measurements. == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 7ksvc4j7nvl0od53ekpjgj2bippbotc 162 161 2023-12-13T13:59:03Z OpenwaterAndrew 3 /* Coherence Length */ 162 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: Figure 1 The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to frequency or wavelength variations over time. Managing chirp is vital for stable speckle signals in our blood flow measurement. == Interferometer Experiment == We used interferometric techniques to compare coherence, frequency, and power stability. The experiment details and sample data are in the repository. == Speckle Contrast Measurements == A direct speckle contrast measurement test assesses coherence length's effect on speckle contrast. High coherence (narrower linewidth) improves blood flow detection. == Effect of Chirping on Speckle Contrast == Figure 2 demonstrates the detrimental effect of chirping on speckle contrast, necessitating stable laser frequency for accurate measurements. == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 1lc555bo4yahuz6qcqcpn7038w4iqjf 163 162 2023-12-13T13:59:32Z OpenwaterAndrew 3 /* Chirp */ 163 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: Figure 1 The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We used interferometric techniques to compare coherence, frequency, and power stability. The experiment details and sample data are in the repository. == Speckle Contrast Measurements == A direct speckle contrast measurement test assesses coherence length's effect on speckle contrast. High coherence (narrower linewidth) improves blood flow detection. == Effect of Chirping on Speckle Contrast == Figure 2 demonstrates the detrimental effect of chirping on speckle contrast, necessitating stable laser frequency for accurate measurements. == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” ozic6qraqhkzn9selh69asiqe9gcu3x 164 163 2023-12-13T13:59:58Z OpenwaterAndrew 3 /* Interferometer Experiment */ 164 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: Figure 1 The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “interferometer_chirp_test_sample_data.mat” file contains the interferometer sample data * The “interferometer_chirp_test.m” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == A direct speckle contrast measurement test assesses coherence length's effect on speckle contrast. High coherence (narrower linewidth) improves blood flow detection. == Effect of Chirping on Speckle Contrast == Figure 2 demonstrates the detrimental effect of chirping on speckle contrast, necessitating stable laser frequency for accurate measurements. == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 7h6bdre8zp8jpfznmtjr74tox4zmzxl 165 164 2023-12-13T14:00:18Z OpenwaterAndrew 3 /* Speckle Contrast Measurements */ 165 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: Figure 1 The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “interferometer_chirp_test_sample_data.mat” file contains the interferometer sample data * The “interferometer_chirp_test.m” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” == Effect of Chirping on Speckle Contrast == Figure 2 demonstrates the detrimental effect of chirping on speckle contrast, necessitating stable laser frequency for accurate measurements. == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” pflbp0gxnv0o91ltfdz5aeatwlv8uyu 167 165 2023-12-13T14:01:52Z OpenwaterAndrew 3 /* Effect of Chirping on Speckle Contrast */ 167 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: Figure 1 The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “interferometer_chirp_test_sample_data.mat” file contains the interferometer sample data * The “interferometer_chirp_test.m” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of laser chirp on speckle]] == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” dznvi9kuleii1ns5h9n845akr4n83dw 168 167 2023-12-13T14:02:11Z OpenwaterAndrew 3 /* Effect of Chirping on Speckle Contrast */ 168 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: Figure 1 The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “interferometer_chirp_test_sample_data.mat” file contains the interferometer sample data * The “interferometer_chirp_test.m” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == - “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” c4al4p1d659k2tyqsxm5hpi15rolcml 169 168 2023-12-13T14:02:27Z OpenwaterAndrew 3 /* References */ 169 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: Figure 1 The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “interferometer_chirp_test_sample_data.mat” file contains the interferometer sample data * The “interferometer_chirp_test.m” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” hleg7kp9h7nlgwy7odg2t8b4iunpdx9 171 169 2023-12-13T14:04:01Z OpenwaterAndrew 3 /* Coherence Length */ 171 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “interferometer_chirp_test_sample_data.mat” file contains the interferometer sample data * The “interferometer_chirp_test.m” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” hcz32a8k9p1krbeh1mqoog5sp97muk0 172 171 2023-12-13T14:16:38Z OpenwaterAndrew 3 /* Coherence Length */ 172 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula (need to add formulas) [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “interferometer_chirp_test_sample_data.mat” file contains the interferometer sample data * The “interferometer_chirp_test.m” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 6dr4bzpdk2ksd3ld3cko5gvucpf3glf 173 172 2023-12-13T14:33:07Z OpenwaterAndrew 3 /* Interferometer Experiment */ 173 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula (need to add formulas) [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” h0togdo4sb8ibr61xvmhosif4xzisqx 185 173 2023-12-13T16:39:07Z OpenwaterAndrew 3 /* Coherence Length */ 185 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” s3f2e0gb4vpux0gp3ygarr54lj5ubnj 215 185 2023-12-13T18:00:54Z Openwaterpete 5 215 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” nxekmzocvrtpf0l550jwsgtmpnaec8d 218 215 2023-12-13T18:16:49Z OpenwaterAndrew 3 /* Speckle Contrast Measurements */ 218 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 4btauxdpc0k5ypeal0p5a6ijqhnf136 219 218 2023-12-13T18:17:02Z OpenwaterAndrew 3 /* Effect of Chirping on Speckle Contrast */ 219 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “interferometer experiment detail” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 8o38lpma3dpm7yd65i16ow8re7mg5ar 323 219 2023-12-13T23:27:13Z 50.227.118.138 /* Interferometer Experiment */ 323 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/main/interferometeric%20characterization.pdf interferometeric charactreization]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 7114hrqes4bufuhyletpvp4v24nvphz 328 323 2023-12-14T00:04:04Z 50.227.118.138 328 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 9f32zkdzf77axm6atz7pn1ppaw1gl59 336 328 2023-12-14T00:24:28Z 50.227.118.138 336 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification WAITING TO EDIT GOOGLE DOC> = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” nqt4lasxztnar7ielhx6x73x4krzpu9 339 336 2023-12-14T00:41:06Z OpenwaterGambhir 10 339 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 9f32zkdzf77axm6atz7pn1ppaw1gl59 340 339 2023-12-14T00:42:00Z OpenwaterGambhir 10 /* Interferometer Experiment */ 340 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” izcq7dzctq1p4drdoja4cz83xeory1m 344 340 2023-12-14T00:44:21Z OpenwaterGambhir 10 /* Interferometer Experiment */ 344 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” hk8nw8lrbq8mrivn1p1hwkqsfjhxw6d 345 344 2023-12-14T00:45:43Z OpenwaterGambhir 10 /* Interferometer Experiment */ 345 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 7iv1t8izs2vzr15y3qhxuxb6o8649tu 347 345 2023-12-14T00:46:54Z OpenwaterGambhir 10 /* Interferometer setup -version one */ 347 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|none|thumb|648x648px|''The Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.]] * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 23zm1cdjuvjem7b0j4b15crpd4otb95 351 347 2023-12-14T00:53:30Z OpenwaterGambhir 10 /* Interferometer setup -version one */ 351 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|none|thumb|648x648px|''The Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.]] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. [[File:InterferometerData1.png|none|thumb|565x565px|''The Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” s99jc72zkb5pxjvqh4xl9bymcms3g7x 352 351 2023-12-14T00:54:01Z OpenwaterGambhir 10 /* Interferometer setup -version one */ 352 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == The purpose of this characterization is to qualify the lasers used in Openwater blood flow devices and ensure that they meet performance requirements. Long coherence length (narrow linewidth) is one of the most critical characteristics of the lasers used in our application. See Gen 1 Blood Flow White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|none|thumb|648x648px|''The Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.]] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. [[File:InterferometerData1.png|none|thumb|565x565px|''The Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 7p817oc1hz67qlve2e913gvfqbm4xbr 353 352 2023-12-14T00:55:24Z OpenwaterGambhir 10 /* Introduction */ 353 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|none|thumb|648x648px|''The Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.]] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. [[File:InterferometerData1.png|none|thumb|565x565px|''The Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 44vk8vwtkv117r7h36ri5n72zxc54er 354 353 2023-12-14T00:55:39Z OpenwaterGambhir 10 /* Laser Characterization and Qualification */ 354 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due the factors : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|none|thumb|648x648px|''The Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.]] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. [[File:InterferometerData1.png|none|thumb|565x565px|''The Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 588ia7cpbfefpke8ybgiqj4evee4wx4 357 354 2023-12-14T01:16:07Z OpenwaterGambhir 10 /* Interferometer setup -version one */ 357 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|none|thumb|691x691px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.]] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. [[File:InterferometerData1.png|none|thumb|565x565px|''The Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|none|thumb|970x970px]] [[File:Simplified interferometer setup.png|none|thumb|551x551px|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals. Figure 4 shows the simplified Michelson interferometer setup.'']] * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” qaznx7tvkyrfjbn91lt6qnreyqgm6ds 358 357 2023-12-14T01:26:23Z OpenwaterGambhir 10 /* Interferometer setup -version one */ 358 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 40vf8ie74d7mh06zksnrud13cx4dnpt 359 358 2023-12-14T01:28:19Z OpenwaterGambhir 10 /* Simplified Michelson interferometer setup */ 359 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === ====Set the position of the adjustable arm of the interferometer.==== ====Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. ==== * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 8ijxtkz3q4po2grt372qdeajmrbocnk 360 359 2023-12-14T01:28:51Z OpenwaterGambhir 10 /* Alignment and optimization of the interferometer */ 360 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === ====Set the position of the adjustable arm of the interferometer.==== ====Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. ==== * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” sdyc03d300qslkxd2iq5ied8e4zcz62 361 360 2023-12-14T01:29:03Z OpenwaterGambhir 10 /* Alignment and optimization of the interferometer */ 361 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === ====Set the position of the adjustable arm of the interferometer.==== ====Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. ==== * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 00c04fmshw1bkwgrnxrnaoonoyf51vv 362 361 2023-12-14T01:30:50Z OpenwaterGambhir 10 /* Alignment and optimization of the interferometer */ 362 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” m963aml83xsiqf1y7j3p60lfeckvvwm 363 362 2023-12-14T01:31:45Z OpenwaterGambhir 10 /* Alignment and optimization of the interferometer */ 363 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === * Set the position of the adjustable arm of the interferometer. * Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. * Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. * Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. * Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 1f3nkavvddwotun4xysi8i0dtx58aq2 364 363 2023-12-14T01:32:09Z OpenwaterGambhir 10 /* Alignment and optimization of the interferometer */ 364 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === <nowiki>*</nowiki> Set the position of the adjustable arm of the interferometer. <nowiki>*</nowiki> Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. <nowiki>*</nowiki> Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. <nowiki>*</nowiki> Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. <nowiki>*</nowiki> Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * Set the position of the adjustable arm of the interferometer. * Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. * Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. * Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. * Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” cyeenw1t827j9jatwpsv7olpxcwh9dc 365 364 2023-12-14T01:32:56Z OpenwaterGambhir 10 /* Alignment and optimization of the interferometer */ 365 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * Set the position of the adjustable arm of the interferometer. * Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. * Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. * Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. * Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” nr7gq5ad30qthf6z1wtjnrstfzydqs8 366 365 2023-12-14T01:33:09Z OpenwaterGambhir 10 /* Alignment and optimization of the interferometer */ 366 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula [[File:Coherence length equation.PNG|300 px]] [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 4hiqosxjry3yja940vug18z64q7zh9f 372 366 2023-12-14T11:32:53Z OpenwaterAndrew 3 /* Coherence Length */ 372 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” b3tp6jagnmdjr7vg7jzukucma46q2v3 407 372 2023-12-14T23:50:04Z 50.227.118.138 /* Interferometer setup -version one */ 407 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. __TOC__ === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” i40a712c6jowdun3sx3nd9s8bascfzf 408 407 2023-12-14T23:51:23Z 50.227.118.138 408 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” invid8gn89m1nw8fkm8q7fd3wt6la3n 411 408 2023-12-15T00:08:24Z Openwaterpete 5 411 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. ![mopa block diagram] (file:///C:/Users/pmher/Downloads/laser%20block%20dia.png) This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” tq00cebze5d3crlgib3dtakq77bfuol 412 411 2023-12-15T00:09:15Z Openwaterpete 5 412 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” invid8gn89m1nw8fkm8q7fd3wt6la3n 413 412 2023-12-15T00:11:16Z Openwaterpete 5 413 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. file:///C:/Users/pmher/Downloads/laser%20block%20dia.png == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” qtmu4myk17olo1sdnjta4bsidk85me9 414 413 2023-12-15T00:11:43Z Openwaterpete 5 414 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” invid8gn89m1nw8fkm8q7fd3wt6la3n 416 414 2023-12-15T00:34:46Z Openwaterpete 5 /* Background */ 416 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == [[File:Laser block dia.png|thumb|MOPA block diagram]] The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” j2j0g92b12bix69gedvguvwubwb8ctq 417 416 2023-12-15T00:44:18Z Openwaterpete 5 /* Background */ 417 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == [[File:Laser block dia.png|thumb|MOPA block diagram]] The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. Laser system In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” ft4d5kmxuux06nmuovxijyob3a4l696 418 417 2023-12-15T00:45:34Z Openwaterpete 5 /* Introduction */ 418 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. ==Laser system== In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Background == [[File:Laser block dia.png|thumb|MOPA block diagram]] The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. Laser system In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” ejj4xx5jv7deyjmav98nl10gwjaly6j 419 418 2023-12-15T00:46:00Z Openwaterpete 5 /* Background */ 419 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. ==Laser system== In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Background == [[File:Laser block dia.png|thumb|MOPA block diagram]] The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 5b0nkjxkqdgxeym2vwshn8qaphizytm 420 419 2023-12-15T00:46:27Z Openwaterpete 5 420 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == [[File:Laser block dia.png|thumb|MOPA block diagram]] The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” by138ueskg9o2t796ydt8totseffb09 422 420 2023-12-15T01:38:57Z Openwaterpete 5 /* Laser system */ 422 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == [[File:Laser block dia.png|thumb|MOPA block diagram]] The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Source mod.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” pzs6ekmz7ptpu5ta90j6lr8ekwggh91 423 422 2023-12-15T01:39:27Z Openwaterpete 5 /* Background */ 423 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Source mod.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” owtqvfxlpxlqywpx6slp5mmjip0tpgh 424 423 2023-12-15T01:39:41Z Openwaterpete 5 /* Laser system */ 424 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|MOPA block diagram]] [[File:Source mod.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. Coherence in light refers to the degree of correlation between different points in a wave or between different waves. In a laser, the light is emitted coherently, meaning the waves have a consistent phase relationship—peaks and troughs of the waves align in a specific way. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 2wn6ykcftvjqufm7ssuf064ax59cp14 425 424 2023-12-15T01:40:13Z Openwaterpete 5 /* Coherence Length */ 425 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|MOPA block diagram]] [[File:Source mod.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” mf6hyub3qecga6pptv8qgr6iez3kcpf 428 425 2023-12-15T11:17:54Z 73.241.144.92 /* Introduction */ 428 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Laser spatial mode test. * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|MOPA block diagram]] [[File:Source mod.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” kecrqcaly33is2b6ymo4bkzzuq96r8j 430 428 2023-12-15T16:41:17Z Openwaterpete 5 /* Laser system */ 430 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Laser spatial mode test. * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|MOPA block diagram]] [[File:Source mod.png|thumb|figure 2 source module]] [[File:Source mod 3.png|thumb|figure 3 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” r6twbl0nohgaxbqs77fnrhz6d214h2q 431 430 2023-12-15T16:56:56Z Openwaterpete 5 431 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required the development of new tests and procedures for characterization and optimization to ensure the lasers used in Openwater blood flow devices and high performance requirements. See Diagnosis White Paper for more detailed background and theory of operation. This document describes testing of lasers used in Openwater systems. It begins with a short overview of important background topics including coherence length and chirp. It then describes the specific tests used to qualify our lasers, namely * Laser spatial mode test. * Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. * Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 3 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 1: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” sc7yc7e5p8x5h7is67q8ikleay3jqe3 433 431 2023-12-15T17:13:49Z Openwaterpete 5 /* Chirp */ 433 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: Laser spatial mode test. Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 3: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Speckle.png|thumb|figure 4]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” omrrdo0xb0yiig8ti7pejdejfvobynn 434 433 2023-12-15T17:15:49Z Openwaterpete 5 /* Introduction */ 434 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 3: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Speckle.png|thumb|figure 4]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 84wg5kdpo28z5vynzv84d7f9kkdy6qd 435 434 2023-12-15T17:19:03Z Openwaterpete 5 /* Introduction */ 435 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 3: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Speckle.png|thumb|figure 4]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” pcn4mtcly33h5drcwwdlcm6z7jz5gsw 436 435 2023-12-15T17:20:08Z Openwaterpete 5 /* Introduction */ 436 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the [Openwater Blood Flow and Stroke Diagnosis Wiki] (http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology). This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 3: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Speckle.png|thumb|figure 4]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” iaazaxhnxtht90d6wpq6xz03b3ajcnr 437 436 2023-12-15T17:20:45Z Openwaterpete 5 /* Introduction */ 437 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 3: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Speckle.png|thumb|figure 4]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. Techniques and devices are employed to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 4b2tx4afsswwgtpkyc03838oejrpmcj 438 437 2023-12-15T18:28:41Z 50.227.118.138 438 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 3: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Speckle.png|thumb|figure 4]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” rv5fg7mf599wz5d06e2xyyn272su4ny 439 438 2023-12-15T18:30:17Z 50.227.118.138 /* Chirp */ 439 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 3: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” ewcs1r3h021zyl4umbbmqvqnax20p1z 441 439 2023-12-15T18:34:36Z OpenwaterGambhir 10 /* Chirp */ 441 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. [[File:Laser spectral width.PNG|thumb|Figure 3: Laser spectral width]] The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 shows the interferometer signals with and without chirp (left) and the corresponding speckle images (right)]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” d3zbetx9wici0wm6ybkkfd9dkaoj3zu 443 441 2023-12-15T18:43:42Z OpenwaterGambhir 10 /* Chirp */ 443 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 6irszybsf1n31b8f663bbg1zwz28fwl 444 443 2023-12-15T18:46:21Z OpenwaterGambhir 10 444 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Improtance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” pannchfpd5qc0v64hngpe8rcn0qk17m 445 444 2023-12-15T18:47:41Z OpenwaterGambhir 10 445 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. * Required test equipment CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM * Experimental setup Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 62onwdmxookhmj78tsz1zwrecf1soi3 446 445 2023-12-15T18:48:38Z OpenwaterGambhir 10 446 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== * Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== * Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” bwpy25jyxu31lm0wkyb20o4q0nmgfkw 448 446 2023-12-15T18:56:50Z OpenwaterGambhir 10 /* * Experimental setup */ 448 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== * Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== * Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” qrjn14z035i38kph0ja9nbhrgjoa8js 449 448 2023-12-15T18:59:20Z OpenwaterGambhir 10 449 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: ===== * Open the DataRay software ==== ===== * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge ==== ===== * Exposure control (right click on “Imager Gain” or “Exposure time”): ==== ====== * Enable auto gain adjustment ====== ====== * Exposure time is not relevant due to external triggering ====== ===== * Make sure to zoom in to 3x for the center image ===== == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” mc6zrhjkmfft8gilr9v3vyxlpb41eja 451 449 2023-12-15T19:05:47Z OpenwaterGambhir 10 451 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 9y38xrpcentfog74oxo52jo4ec0wah6 453 451 2023-12-15T19:08:03Z OpenwaterGambhir 10 /* Experimental setup */ 453 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 5 spatial mode of a laser with good performance]] == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 4bhkvf3lkjwvn8oxzznp8ocrq14wnxg 454 453 2023-12-15T19:11:55Z OpenwaterGambhir 10 /* Chirp */ 454 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 5 spatial mode of a laser with good performance]] == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 72j06mzkqkej6pmml3jhhuxkampppob 455 454 2023-12-15T19:13:47Z OpenwaterGambhir 10 455 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 5 spatial mode of a laser with good performance]] First beam profile was captured using a laser with good performance. As can be seen in Figure 5 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” azd9wfrigquk13bxk0dct04o52p7wiz 459 455 2023-12-15T19:21:53Z OpenwaterGambhir 10 459 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 spatial mode of a laser with good performance|424x424px]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 775c6bhqe56v4v9s7x6dfbvg7nijygp 460 459 2023-12-15T19:22:27Z OpenwaterGambhir 10 460 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 spatial mode of a laser with good performance|424x424px]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Experiment == We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === Introduction === An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. === Interferometer setup -version one === The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michaelson interferometer is devised as shown in the Figure (1) [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” ehusp6cgtyw6o44ge2lwtyppkdt2mx4 465 460 2023-12-15T19:33:44Z OpenwaterGambhir 10 465 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 Good performance, single central lobe|424x424px]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8 [[File:Interferometer setup 1.png|thumb|964x964px|''Figure1 shows the original home built interferometer setup for the characterization of the laser, The PD3 monitors the intensity, the PD1 and PD2 collect the interferometer signal in two orthogonal directions to monitor the phase noise''.|border|left]] [[File:InterferometerData1.png|border|thumb|410x410px|''Figure 2a and 2d show the interferometer signals collected at different PZT positions for good and bad lasers respectively. The Figure 2b and 2e are the intensity mean per pulse as a function of the PZT position and the Figure 2c and Figure 2f show the sample of phase variation during the laser pulse for single PZT position.'']] The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” ssnxf0av2k4pogu2p1gew552frxfeh2 467 465 2023-12-15T19:36:49Z OpenwaterGambhir 10 467 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 Good performance, single central lobe|424x424px]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|569x569px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 0pp33lbrficrvypw14le9e3czo00624 468 467 2023-12-15T19:38:22Z OpenwaterGambhir 10 468 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 Good performance, single central lobe|424x424px]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|648x648px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” aue4b0w3rib3vmz3e9ra08xsyy7kyk1 469 468 2023-12-15T19:38:43Z OpenwaterGambhir 10 469 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 Good performance, single central lobe|424x424px]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|648x648px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 8tifkulzy38cced6nynjjrkytkzw3ga 470 469 2023-12-15T19:42:46Z OpenwaterGambhir 10 470 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 Good performance, single central lobe|424x424px]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|648x648px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” tpy9cwq6qy1ykf21xttcl4pl3gndmxw 472 470 2023-12-15T19:46:55Z OpenwaterGambhir 10 472 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 Good performance, single central lobe|424x424px]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|648x648px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” bq9q861x19htpn3n3njomj5xvb0y0br 473 472 2023-12-15T19:47:39Z OpenwaterGambhir 10 473 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 Good performance, single central lobe|424x424px]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|648x648px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” kstk80rzym1uhud5zoieyzpqzxvr23f 474 473 2023-12-15T19:50:49Z 69.181.108.2 /* Experimental setup */ 474 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 Good performance, single central lobe|424x424px]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|648x648px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” dp6hq5r9bcbocgk8x1189w3l6hwdmpg 475 474 2023-12-15T19:51:21Z OpenwaterGambhir 10 475 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 Good performance, single central lobe|424x424px]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|648x648px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” f5g12npnnxwxi3pj2i48c064nk0eygs 476 475 2023-12-15T19:51:34Z 69.181.108.2 /* Experimental setup */ 476 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|left|thumb|Figure 6 Good performance, single central lobe]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|648x648px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 2xc6jbg3177w8k9866xo0lld62zvms8 477 476 2023-12-15T19:52:44Z 69.181.108.2 /* Experimental setup */ 477 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|Figure 6 Good performance,|448x448px single central lobe]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|648x648px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” tj8zqknbfri6nvkaaqvhxjponl8nd6w 480 477 2023-12-15T19:53:28Z 69.181.108.2 /* Experimental setup */ 480 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb||448x448px|Figure 6 Good performance, single central lobe]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. [[File:Simplified interferometer.png|thumb|778x778px|left|''Figure 3 displays the major components used to build the interferometer and the instruments used to drive PZT, and monitor and collect the interferometer signals.'']] [[File:Simplified interferometer setup.png|none|thumb|648x648px|''Figure 4 shows the simplified Michelson interferometer setup.'']] === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” r1w13m69s4f4ufesoiujx54yhj0jqzt 481 480 2023-12-15T19:54:38Z 69.181.108.2 /* Simplified Michelson interferometer setup */ 481 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb||448x448px|Figure 6 Good performance, single central lobe]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 4kfrfhfa9bdqit956zq0detzwptj7ym 483 481 2023-12-15T19:58:45Z OpenwaterGambhir 10 483 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). The output of the tapered amplifier (TA) and optical isolator is split using (8/92) beam splitter cube (BS). The 8% beam is aligned to the photodetector (PD3) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected at a shorter, fixed distance while the other is reflected at a longer, adjustable delay with a glass retro-reflector mounted onto the piezo-stage. The delay is set to ~100 cm in the longer arm of the interferometer. It is then combined with the reflected beam from the shorter, fixed arm. The combined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and they are aligned to the photodetectors (PD1 and PD2) as shown in the Figure 1 to obtain the phase information along with the coherence behavior of the laser. The typical interferometer signals and the phasor diagrams with the TA driven at a current of 5A with pulse width of 100 us are shown in Figure 2 for the comparison between good and underperforming TAs. A well performing TA shows smooth variation in the interference as we sweep the position of piezo (PZT) and the phase noise is tightly scattered compared to the underperforming TA. A poorly performing TA shows broader scattering of the phase noise and the interference pattern is also washed out. Figure 2b and 2e depict the mean interferometer signal across the pulse. The interferometer contrast is significantly reduced for the underperforming laser shown in Figure 2e compared to that of the well-performing laser in Figure 2b. === Simplified Michelson interferometer setup === The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 3. These are available in Thorlabs with the path number shown below each component. The test setup is shown in Figure 4. === Alignment and optimization of the interferometer === 1) Set the position of the adjustable arm of the interferometer. 2) Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. 3) Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 5a. 4) Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 5a. 5) Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 5b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT) (Note: The pattern of the fringe depends upon the alignments) * The “[[Interferometric Characterization|interferometeric charactreization]]” file contains the information about the optical setup for the interferometer, the alignment procedure for the testing and result plots. * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test_sample_data.mat interferometer_chirp_test_sample_data.mat]” file contains the interferometer sample data * The “[https://github.com/OpenwaterHealth/opw_laser/blob/fe3482ed9b6710693255e4088df71230fdf5e5f8/interferometer_chirp_test.m interferometer_chirp_test.m]” file contains the script that analyzes the sample test data and displays the results. == Speckle Contrast Measurements == After characterization of the laser system in the interferometric setup, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined in detail in the file “Speckle Contrast Measurements” [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” cya9genxnjipo992zkbqnjswaoindq2 485 483 2023-12-15T20:21:12Z OpenwaterGambhir 10 485 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” l6hsgds6kuowlan315a5e545ovbqw1r 486 485 2023-12-15T20:22:45Z OpenwaterGambhir 10 486 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” higskwnraj6g2gxwiaogmofw4rh1ozk 487 486 2023-12-15T20:24:11Z OpenwaterGambhir 10 487 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” e8z6gty15gpe1660kpl0267600iu28z 488 487 2023-12-15T20:27:53Z OpenwaterGambhir 10 488 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal. # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown below in table 1. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 9q9en3pjjii8pvw8j4ylfl7ihibiuk6 490 488 2023-12-15T20:31:10Z OpenwaterGambhir 10 490 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:Pathdifference1.png|left|thumb|512x512px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal. # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown below in table 1. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” p0gktyhi7gi5j1nucfxcbgexiocej9z 494 490 2023-12-15T20:44:25Z OpenwaterGambhir 10 /* Calculation of the path difference: */ 494 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|633x633px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal. # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown below in table 1. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 0cafk7kxz3i2de06fymzh4h60me0731 497 494 2023-12-15T21:00:33Z OpenwaterGambhir 10 497 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|717x717px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” rmg8jb89xcyrydnw8j356erpgeiumkg 498 497 2023-12-15T21:01:45Z OpenwaterGambhir 10 498 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|653x653px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” qt2poujsxwhscvvzmk8np8wia569g3k 499 498 2023-12-15T21:02:21Z OpenwaterGambhir 10 499 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|664x664px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 6fpaj7idky8gcsrwx7yruwd4ofrfzi0 500 499 2023-12-15T21:02:55Z OpenwaterGambhir 10 500 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|684x684px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” g3auqn1io75d7msvehb69nfsjgsc5ey 501 500 2023-12-15T21:03:07Z OpenwaterGambhir 10 /* Calculation of the path difference: */ 501 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|684x684px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” fyhz4g6prmfgndi5efmm469phh4oml5 502 501 2023-12-15T21:03:58Z OpenwaterGambhir 10 502 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 43qw597g3hldsvt5zmvq6fcakhdxxtk 503 502 2023-12-15T21:12:36Z OpenwaterGambhir 10 503 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> We define the path difference as <math>path \ difference = 2(x_2 - x_1) = \frac{\lambda^2}{\Delta\lambda}</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 7iy9fho05s7sd6364ajeujnjupilgcs 504 503 2023-12-15T21:15:36Z OpenwaterGambhir 10 504 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> We define the path difference as <math>path \ difference = 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. <math>path \ difference = \frac{\lambda^2}{\Delta\lambda}</math> [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” n946am9pnn3m7h8ff6gr1uhnljhimei 505 504 2023-12-15T21:24:30Z OpenwaterGambhir 10 505 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as <math>path \ difference = 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. <math>path \ difference = \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math>x_{pos0}<math> equals to ~ 452.35 mm. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 9v9lu4ga6qn8cc4rqgdgslkxf4lxdoy 510 505 2023-12-15T21:54:58Z OpenwaterGambhir 10 510 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 5sf48zqe4deo6pz9bvjhdi1tzx2231z 511 510 2023-12-15T21:58:51Z OpenwaterGambhir 10 511 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 2s5vdnx8mzxa7sq7e0c81lmb1y4ituw 513 511 2023-12-15T22:13:21Z OpenwaterGambhir 10 513 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m [[File:FFS simulated.png|thumb|720x720px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” lp2zotg9llz1u3of9hqc609ql6utgvx 516 513 2023-12-15T22:23:36Z OpenwaterGambhir 10 516 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as path difference <math>= 2(x_2 - x_1)</math> [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]]Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. [[File:Chirp interferometer signal1.png|thumb|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] [[File:Chirp interferometer signal2.png|thumb|Figure 14b]] [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 3ju749nky05pqgqqsfj94z2uwyh9yq6 517 516 2023-12-15T22:30:06Z OpenwaterGambhir 10 517 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as path difference <math>= 2(x_2 - x_1)</math> [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]]Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. [[File:Chirp interferometer signal1.png|thumb|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] [[File:Chirp interferometer signal2.png|thumb|Figure 14b]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 54b [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 2qb1ry8yfjnoyj0gvp5ys26irgby8jz 524 517 2023-12-15T22:49:38Z OpenwaterGambhir 10 524 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|621x621px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b [[File:Chrip plot1.png|none|thumb|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] [[File:Chirp plot2.png|none|thumb|Figure 15b]] [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” qbqmrq4e1amyefve2xs7x2pq9i7g993 526 524 2023-12-15T22:55:33Z OpenwaterGambhir 10 526 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|791x791px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|789x789px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” k3tn0qh73c77nrhuaxwxx12feni3uka 527 526 2023-12-15T22:59:39Z OpenwaterGambhir 10 527 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|791x791px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|789x789px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature Ths . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” nfbyxh7r43gx50ss54o0pokfgst3b2q 529 527 2023-12-15T23:02:26Z OpenwaterGambhir 10 529 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature Ths . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” jntw6nokgk9f3x68xl7jbn6okrlotd9 530 529 2023-12-15T23:06:05Z OpenwaterGambhir 10 530 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 6mouxdx6ma7jqktzsvyomnyhxn3734m 531 530 2023-12-15T23:07:17Z OpenwaterGambhir 10 531 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” ixoch23i8x40b8y9uahv3lqcmh3rp8e 534 531 2023-12-15T23:11:03Z OpenwaterGambhir 10 534 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” cyrao8p9qhtggq19gwejsq98qqyu2sa 537 534 2023-12-15T23:12:50Z OpenwaterGambhir 10 537 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 90g2woznes5xpw4rtecm87nh7wxkpz8 539 537 2023-12-15T23:20:08Z OpenwaterGambhir 10 539 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 81hy3qdn62iejlcgaj2e7n6mthxgatl 541 539 2023-12-15T23:22:24Z OpenwaterGambhir 10 541 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|647x647px|Figure 18 Hardware components used in speckle measurement setup]] [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == Effect of Chirping on Speckle Contrast == The detrimental effect of the chirping is shown in Figure 2. The speckle contrast is significantly washed out due to the presence of the chirping (or oscillations in the interferometer signal) during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” tuv9nxfqqhv8jg62uyrrs5vhd7etzgu 544 541 2023-12-15T23:25:33Z OpenwaterGambhir 10 544 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|647x647px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 9pze8xbob4l0rno72on5no0phgt30m4 547 544 2023-12-15T23:28:40Z OpenwaterGambhir 10 547 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|449x449px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|793x793px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” dzdvdrmsej25vpd0ke5ioqjleyqu0u9 548 547 2023-12-15T23:35:25Z OpenwaterGambhir 10 548 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|449x449px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|793x793px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ### Basic Settings: #### No downsampling, choose 1(4112x3008) option #### From the next dropdown menu select “Raw16” option #### Do NOT select “Auto exposure” option #### Set the exposure time to 10ms #### On Gain selector, select Analog #### Set Gain at 0dB #### ROI set on, defined as following (full-frame will not work properly at 30 Hz): ##### Offset X: 1032 ##### Offset Y: 752 ##### Width: 2056 ##### Height: 1504 [[File:Chirp speckle effect.PNG|thumb|Figure 2: Effect of chirp on laser speckle contrast]] == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” 2ski4vl5m0j72ea3dqea2t46yd2ktuo 549 548 2023-12-15T23:46:07Z OpenwaterGambhir 10 /* XIMEA camera set-up and software settings: */ 549 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|449x449px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|793x793px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (shown below). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (shown below). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” bwtidua2w72zuzdip05w8wovgzqqhje 550 549 2023-12-15T23:46:46Z OpenwaterGambhir 10 /* References */ 550 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|449x449px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|793x793px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (shown below). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (shown below). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == # “Time resolved reflectance and transmittance for non-invasive measurements of tissue optical properties, Michael S. Patterson, B.Chance and B.C. Wilson” #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” iacn8kmzptxcorppu1k6ehc7m3azb83 552 550 2023-12-15T23:51:20Z OpenwaterGambhir 10 552 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|449x449px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|793x793px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (shown below). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (shown below). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” q6267hqzzmcuu5b8pejg9l16litkq4q 561 552 2023-12-16T00:14:24Z OpenwaterGambhir 10 561 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|449x449px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|793x793px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SampleScriptAndResults.jpg|thumb|258x258px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” kgbgacrmari75ltw7jewpursizk8rmx 566 561 2023-12-16T00:34:55Z OpenwaterGambhir 10 566 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|448x448px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|448x448px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== [[File:Chirp inteferometer12.png|thumb|683x683px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. ==== Measurement of the chirping of the laser during pulse: ==== [[File:Chirp plot12.png|thumb|677x677px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|669x669px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|449x449px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|793x793px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|303x303px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” s8q7dqui1o1m27lmawp9bd9gkya0b7n 574 566 2023-12-16T19:05:12Z 73.241.144.92 574 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|370x370px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” 9fi0vdzcl4ty5c9c8ajaeg70mxccqxd 575 574 2023-12-16T19:06:21Z 73.241.144.92 /* Experimental setup */ 575 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|334x334px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror mount edge position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” mgbjchwoef3tovu2nld47sndkygo02j 576 575 2023-12-16T19:07:59Z 73.241.144.92 /* Calculation of the path difference: */ 576 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|334x334px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Speckle1.png|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” j8k6h1godt6rrepnnzp33jo2jge6b2p 580 576 2023-12-17T00:24:55Z OpenwaterGambhir 10 580 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|334x334px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Ezgif.com-cut.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” oegowupa8htuap74h3tfxsmwkup2ilf 581 580 2023-12-17T00:31:49Z OpenwaterGambhir 10 581 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|334x334px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” pqdl1tyjkxnht5elkzjv0pp2lkfmnc7 582 581 2023-12-17T22:17:42Z 73.241.144.92 /* Calculation of the full fringe shift range (FFS) for the setup: */ 582 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|334x334px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” 9b1ax99lp6wgyap65fihe9m29j7nwrt 583 582 2023-12-17T22:18:29Z OpenwaterGambhir 10 /* Calculation of the full fringe shift range (FFS) for the setup: */ 583 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|334x334px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” aa3zbv31ibn4p2qnnqqwhh1jv28f4mu 584 583 2023-12-17T22:19:59Z OpenwaterGambhir 10 /* Measurement of the chirping of the laser during pulse: */ 584 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|334x334px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can have mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” ck8vxlk9ai7wsn90rze3ygkf03uv951 585 584 2023-12-17T22:20:54Z OpenwaterGambhir 10 /* Measurement of the chirping of the laser during pulse: */ 585 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|334x334px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” k6n6d71ywr66xuc7z6027u3lvprj4qm 586 585 2023-12-17T22:22:07Z OpenwaterGambhir 10 /* Speckle Contrast Measurements */ 586 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|334x334px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:Ezgif.com-cut.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” cpx8te5gnpme4r5aawjfhtku1mtke27 587 586 2023-12-17T22:22:37Z OpenwaterGambhir 10 /* Speckle Contrast Measurements */ 587 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|334x334px|Figure 7 Poor performance: many spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” k6n6d71ywr66xuc7z6027u3lvprj4qm 589 587 2023-12-17T23:03:15Z OpenwaterGambhir 10 /* Experimental setup */ 589 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|Figure 7 Poor performance: multiple spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of ΔTj ≤ 2E-3 oC within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” i74lrcpix2pndmeduk150eep5lzojba 590 589 2023-12-17T23:06:21Z OpenwaterGambhir 10 /* Conclusion: */ 590 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|Figure 7 Poor performance: multiple spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of Δ𝝀 equals to ~0.223 nm corresponding to 77 GHz. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by ~ 2.6oC. The amount of acceptable chirping is <65 MHz. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> ΔTj ≤ 2E-3^o</math>C within the pulse width ≥ 250µs. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” rabiy6n6qci0lmbmdt572msshzocn8f 591 590 2023-12-17T23:24:28Z OpenwaterGambhir 10 591 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|figure 1 MOPA block diagram]] [[File:Source mod 3.png|thumb|figure 2 source module]] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|Figure 3 Laser Lineshape]] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|Figure 5 Spatial mode test setup]] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|Figure 6 Good performance, single central lobe|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|Figure 7 Poor performance: multiple spatial modes]] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|Figure 8 A Michelson interferometer setup used for laser chirp measurements]] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals. ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT.'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8.'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position.'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference.]] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero.]] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively.]] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively.]] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time.]] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|Figure 16]] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of <math> \Delta \lambda \approx 0.223 nm </math> corresponding to <math> \approx 77 GHz</math>. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by <math>\approx 2.6^oC</math>. The amount of acceptable chirping is <math> \leq 65 MHz</math>. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> \Delta T_{j} \leq 2E-3 ^oC</math> within the pulse width <math> \leq 250 \mu s</math>. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17.]] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|Figure 18 Hardware components used in speckle measurement setup]] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|Figure 20 shows trigger source to sync the Ximea camera and laser pulse]]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure.]]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|Figure 23 Scripts and results display in the output console]]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” aqdapfevsnnfkronnrn1lnwl1ejxmxg 592 591 2023-12-18T05:29:49Z 73.241.144.92 592 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|''Figure 1 MOPA block diagram'']] [[File:Source mod 3.png|thumb|''Figure 2 source module'']] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|''Figure 3 Laser Lineshape'']] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|''Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)''|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|''Figure 5 Spatial mode test setup'']] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|''Figure 6 Good performance, single central lobe''|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|''Figure 7 Poor performance: multiple spatial modes'']] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|''Figure 8 A Michelson interferometer setup used for laser chirp measurements'']] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|''Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals'' ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|''Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference'']] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|''Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero'']] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|''Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively'']] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|''Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively'']] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|''Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time'']] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|''Figure 16a wavelength versus waste power for different heat sink temperatures, Figure 16b shows relation between junction temperatures and the wavelength of the output laser'']] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of <math> \Delta \lambda \approx 0.223 nm </math> corresponding to <math> \approx 77 GHz</math>. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by <math>\approx 2.6^oC</math>. The amount of acceptable chirping is <math> \leq 65 MHz</math>. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> \Delta T_{j} \leq 2E-3 ^oC</math> within the pulse width <math> \leq 250 \mu s</math>. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17 speckle pattern ]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|''Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17'']] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|''Figure 18 Hardware components used in speckle measurement setup'']] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|''Figure 20 shows trigger source to sync the Ximea camera and laser pulse'']]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Freamerate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|''Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure'']]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|''Figure 23 Scripts and results display in the output console'']]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” dkcabclc0zdnp0n6t52e62y28snq8j8 593 592 2023-12-18T17:15:34Z 73.241.144.92 /* XIMEA camera set-up and software settings: */ 593 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|''Figure 1 MOPA block diagram'']] [[File:Source mod 3.png|thumb|''Figure 2 source module'']] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|''Figure 3 Laser Lineshape'']] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|''Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)''|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|''Figure 5 Spatial mode test setup'']] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|''Figure 6 Good performance, single central lobe''|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|''Figure 7 Poor performance: multiple spatial modes'']] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|''Figure 8 A Michelson interferometer setup used for laser chirp measurements'']] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|''Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals'' ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|''Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference'']] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|''Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero'']] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|''Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively'']] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|''Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively'']] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|''Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time'']] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|''Figure 16a wavelength versus waste power for different heat sink temperatures, Figure 16b shows relation between junction temperatures and the wavelength of the output laser'']] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of <math> \Delta \lambda \approx 0.223 nm </math> corresponding to <math> \approx 77 GHz</math>. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by <math>\approx 2.6^oC</math>. The amount of acceptable chirping is <math> \leq 65 MHz</math>. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> \Delta T_{j} \leq 2E-3 ^oC</math> within the pulse width <math> \leq 250 \mu s</math>. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17 speckle pattern ]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|''Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17'']] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|''Figure 18 Hardware components used in speckle measurement setup'']] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|''Figure 20 shows trigger source to sync the Ximea camera and laser pulse'']]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Frame rate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|''Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure'']]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|''Figure 23 Scripts and results display in the output console'']]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” d5y2yxdzm9i3dblgp4j8mxcp97od6kj 594 593 2023-12-18T17:27:42Z OpenwaterGambhir 10 /* Measurement of junction temperature: */ 594 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|''Figure 1 MOPA block diagram'']] [[File:Source mod 3.png|thumb|''Figure 2 source module'']] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|''Figure 3 Laser Lineshape'']] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|''Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)''|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|''Figure 5 Spatial mode test setup'']] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|''Figure 6 Good performance, single central lobe''|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|''Figure 7 Poor performance: multiple spatial modes'']] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|''Figure 8 A Michelson interferometer setup used for laser chirp measurements'']] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|''Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals'' ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|''Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference'']] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|''Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero'']] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|''Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively'']] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|''Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively'']] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|''Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time'']] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|''Figure 16a wavelength versus waste power for different heat sink temperatures, Figure 16b shows relation between junction temperatures and the wavelength of the output laser'']] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of <math> \Delta \lambda \approx 0.223 nm </math> corresponding to <math> \approx 77 GHz</math>. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by <math>\approx 2.6^oC</math>. The amount of acceptable chirping is <math> \leq 65 MHz</math>. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> \Delta T_{j} \leq 2E-3 ^oC</math> within the pulse width <math> \geq 250 \mu s</math>. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17 speckle pattern ]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|''Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17'']] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|''Figure 18 Hardware components used in speckle measurement setup'']] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|''Figure 20 shows trigger source to sync the Ximea camera and laser pulse'']]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Image below shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Frame rate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|''Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure'']]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|''Figure 23 Scripts and results display in the output console'']]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” gvyyqsfj24dg74akktqsio8nx617mjm 598 594 2023-12-18T21:27:32Z 50.227.118.138 /* XIMEA camera set-up and software settings: */ 598 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|''Figure 1 MOPA block diagram'']] [[File:Source mod 3.png|thumb|''Figure 2 source module'']] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|''Figure 3 Laser Lineshape'']] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|''Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)''|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|''Figure 5 Spatial mode test setup'']] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|''Figure 6 Good performance, single central lobe''|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|''Figure 7 Poor performance: multiple spatial modes'']] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|''Figure 8 A Michelson interferometer setup used for laser chirp measurements'']] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|''Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals'' ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|''Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference'']] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|''Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero'']] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|''Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively'']] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|''Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively'']] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|''Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time'']] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|''Figure 16a wavelength versus waste power for different heat sink temperatures, Figure 16b shows relation between junction temperatures and the wavelength of the output laser'']] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of <math> \Delta \lambda \approx 0.223 nm </math> corresponding to <math> \approx 77 GHz</math>. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by <math>\approx 2.6^oC</math>. The amount of acceptable chirping is <math> \leq 65 MHz</math>. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> \Delta T_{j} \leq 2E-3 ^oC</math> within the pulse width <math> \geq 250 \mu s</math>. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17 speckle pattern ]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|''Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17'']] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|''Figure 18 Hardware components used in speckle measurement setup'']] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|''Figure 20 shows trigger source to sync the Ximea camera and laser pulse'']]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # Figure 21 shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Frame rate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|''Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure'']]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|''Figure 23 Scripts and results display in the output console'']]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” i9cxglh9q0jikh0wbmhcgbk8h1nxnv7 599 598 2023-12-18T21:30:08Z OpenwaterGambhir 10 599 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|''Figure 1 MOPA block diagram'']] [[File:Source mod 3.png|thumb|''Figure 2 source module'']] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|''Figure 3 Laser Lineshape'']] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|''Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)''|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|''Figure 5 Spatial mode test setup'']] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|''Figure 6 Good performance, single central lobe''|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|''Figure 7 Poor performance: multiple spatial modes'']] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|''Figure 8 A Michelson interferometer setup used for laser chirp measurements'']] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|''Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals'' ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|''Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference'']] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|''Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero'']] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|''Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively'']] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|''Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively'']] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|''Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time'']] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|''Figure 16a wavelength versus waste power for different heat sink temperatures, Figure 16b shows relation between junction temperatures and the wavelength of the output laser'']] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of <math> \Delta \lambda \approx 0.223 nm </math> corresponding to <math> \approx 77 GHz</math>. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by <math>\approx 2.6^oC</math>. The amount of acceptable chirping is <math> \leq 65 MHz</math>. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> \Delta T_{j} \leq 2E-3 ^oC</math> within the pulse width <math> \geq 250 \mu s</math>. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17 speckle pattern ]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|''Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17'']] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|''Figure 18 Hardware components used in speckle measurement setup'']] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|''Figure 20 shows trigger source to sync the Ximea camera and laser pulse'']]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # [[File:Speckle sync.png|thumb|''Figure 21 settings for triggering the XIMEA camera'']]Figure 21 shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Frame rate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|''Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure'']]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|''Figure 23 Scripts and results display in the output console'']]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” aizvgk2k56zwsjawsa69xda3cj4eb32 601 599 2023-12-18T21:36:37Z OpenwaterGambhir 10 601 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|''Figure 1 MOPA block diagram'']] [[File:Source mod 3.png|thumb|''Figure 2 source module'']] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|''Figure 3 Laser Lineshape'']] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|''Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)''|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|''Figure 5 Spatial mode test setup'']] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|''Figure 6 Good performance, single central lobe''|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|''Figure 7 Poor performance: multiple spatial modes'']] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|''Figure 8 A Michelson interferometer setup used for laser chirp measurements'']] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|''Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals'' ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|''Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference'']] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|''Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero'']] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|''Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively'']] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|''Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively'']] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|''Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time'']] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|''Figure 16a wavelength versus waste power for different heat sink temperatures, Figure 16b shows relation between junction temperatures and the wavelength of the output laser'']] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of <math> \Delta \lambda \approx 0.223 nm </math> corresponding to <math> \approx 77 GHz</math>. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by <math>\approx 2.6^oC</math>. The amount of acceptable chirping is <math> \leq 65 MHz</math>. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> \Delta T_{j} \leq 2E-3 ^oC</math> within the pulse width <math> \geq 250 \mu s</math>. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17 speckle pattern ]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|''Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17'']] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|''Figure 18 Hardware components used in speckle measurement setup'']] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|''Figure 20 shows trigger source to sync the Ximea camera and laser pulse'']]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings below) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # [[File:XimeaTrig.png|thumb|''Figure 21 settings for triggering the XIMEA camera'']]Figure 21 shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Frame rate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|''Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure'']]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|''Figure 23 Scripts and results display in the output console'']]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” apailmd5d9i6dkf7xvifa474dl04zqw 602 601 2023-12-18T21:38:29Z OpenwaterGambhir 10 /* XIMEA camera set-up and software settings: */ 602 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|''Figure 1 MOPA block diagram'']] [[File:Source mod 3.png|thumb|''Figure 2 source module'']] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|''Figure 3 Laser Lineshape'']] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|''Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)''|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|''Figure 5 Spatial mode test setup'']] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|''Figure 6 Good performance, single central lobe''|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|''Figure 7 Poor performance: multiple spatial modes'']] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|''Figure 8 A Michelson interferometer setup used for laser chirp measurements'']] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|''Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals'' ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|''Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference'']] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|''Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero'']] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|''Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively'']] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|''Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively'']] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|''Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time'']] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|''Figure 16a wavelength versus waste power for different heat sink temperatures, Figure 16b shows relation between junction temperatures and the wavelength of the output laser'']] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of <math> \Delta \lambda \approx 0.223 nm </math> corresponding to <math> \approx 77 GHz</math>. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by <math>\approx 2.6^oC</math>. The amount of acceptable chirping is <math> \leq 65 MHz</math>. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> \Delta T_{j} \leq 2E-3 ^oC</math> within the pulse width <math> \geq 250 \mu s</math>. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17 speckle pattern ]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|''Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17'']] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|''Figure 18 Hardware components used in speckle measurement setup'']] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|''Figure 20 shows trigger source to sync the Ximea camera and laser pulse'']]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings in Figure 21, channel 1) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # [[File:XimeaTrig.png|thumb|''Figure 21 settings for triggering the XIMEA camera'']]Figure 22 shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Frame rate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|''Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure'']]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|''Figure 23 Scripts and results display in the output console'']]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” 7jmzy6j6efe4ivcf4z681xqz7f6w23d 603 602 2023-12-18T21:39:09Z OpenwaterGambhir 10 603 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|''Figure 1 MOPA block diagram'']] [[File:Source mod 3.png|thumb|''Figure 2 source module'']] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|''Figure 3 Laser Lineshape'']] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|''Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)''|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|''Figure 5 Spatial mode test setup'']] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|''Figure 6 Good performance, single central lobe''|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|''Figure 7 Poor performance: multiple spatial modes'']] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|''Figure 8 A Michelson interferometer setup used for laser chirp measurements'']] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|''Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals'' ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|''Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference'']] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|''Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero'']] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|''Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively'']] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|''Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively'']] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|''Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time'']] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|''Figure 16a wavelength versus waste power for different heat sink temperatures, Figure 16b shows relation between junction temperatures and the wavelength of the output laser'']] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of <math> \Delta \lambda \approx 0.223 nm </math> corresponding to <math> \approx 77 GHz</math>. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by <math>\approx 2.6^oC</math>. The amount of acceptable chirping is <math> \leq 65 MHz</math>. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> \Delta T_{j} \leq 2E-3 ^oC</math> within the pulse width <math> \geq 250 \mu s</math>. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17 speckle pattern ]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|''Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17'']] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|''Figure 18 Hardware components used in speckle measurement setup'']] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|''Figure 20 shows trigger source to sync the Ximea camera and laser pulse'']]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings in Figure 21, channel 1) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # [[File:XimeaTrig.png|thumb|''Figure 21 ch1 settings for triggering the XIMEA camera'']]Figure 22 shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Frame rate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|''Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure'']]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|''Figure 23 Scripts and results display in the output console'']]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” 9e92uvx0ocmyb9axissnj0rl7kks2kh 630 603 2023-12-19T18:42:03Z Admin 1 Protected "[[Laser Characterization]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 603 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_laser opw_laser GitHub repository] for the complete collection of files referenced on this page.''' __TOC__ = Laser Characterization and Qualification = == Introduction == Lasers are a critical component of the Openwater blood flow measurement systems. In order to enable blood flow measurements with extremely high sensitivity, we developed new laser technology and custom lasers. In particular, our novel lasers deliver short pulses of long coherence light with very high instantaneous energy in a portable form factor. Developing these lasers required new tests and procedures for characterization and optimization. Our testing ensures the lasers used in Openwater blood flow devices meet our high performance requirements. For more information about how the lasers are used see the Openwater Blood Flow and Stroke Diagnosis Wiki [http://162.246.254.83/index.php/Openwater_Stroke_Diagnosis_Technology]. This document describes many of the laser tests done at Openwater. It begins with a short overview of important background topics including coherence length and chirp. It then describes three important tests used to qualify our lasers, namely: *Laser spatial mode test. *Interferometric characterization to realize the frequency stability or perseverance of the linewidth during the pulse. *Speckle contrast measurements. == Background == The most difficult part of producing the optical illumination for detecting blood flow is achieving a high power illumination source with the required linewidth or coherence length. In order to keep total laser illumination to an acceptably low level per IEC 60825-1 while achieving high peak power, we operate the laser in a pulsed mode. Producing a stable linewidth is much more difficult when operating in this high-current short pulse mode, as opposed to continuous wave (CW) operation. Before discussing the detailed tests used to qualify and characterize lasers for this application, a short summary of coherence length and chirp is presented. ==Laser system == [[File:Laser block dia.png|thumb|''Figure 1 MOPA block diagram'']] [[File:Source mod 3.png|thumb|''Figure 2 source module'']] In our system, we employ what is commonly called a Master Oscillator Power Amplifier (MOPA) laser system. In the first stage, a master oscillator (sometimes called a seed laser) generates a highly coherent laser beam. This optical illumination is then amplified by the power amplifier, which increases the power of the laser with the desired pulse shape and length. This technology enables precise control over the pulse width and frequency, which is crucial for our application. In a typical laser system, pulsed operation is obtained by pulsing the current to the laser cavity. For short pulses the laser cavity does not have time to stabilize resulting in phase and wavelength fluctuations. In our system, the seed laser is run at low power in continuous mode, and as a result emits stable narrow bandwidth light. We have built systems using both volume holographic grating stabilized lasers as well as distributed feedback lasers. High power pulses are obtained by pulsing the current to the amplification stage. We use a tapered amplifier (TA) in our system due to its ability to output high power. == Coherence Length == The coherence length of a laser refers to the spatial extent over which the electromagnetic waves produced by the laser maintain a consistent phase relationship. In simpler terms, it measures how far the light waves remain in sync or coherent. The longer the coherence length, the larger the pathlength difference can be between beams of light emitted from a laser and still show interference effects. As light travels through the body, it scatters and travels along many different paths of different lengths before leaving the body and being detected. If the coherence length is shorter than the path difference between the different light paths, the interference pattern becomes less distinct, and it becomes harder to measure changes in the interference pattern due to blood flow. As a result, we want the coherence length of our lasers to be much longer than these path length differences, which can be up to about 1 meter long. [[File:Laser spectralwidth.png|thumb|''Figure 3 Laser Lineshape'']] The coherence length is related to the spectral width or range of wavelengths present in the laser light. It's calculated using the formula: <math>coherence \ length= \frac{\lambda_0}{\Delta\lambda}</math> where <math>\lambda_0</math> is the central wavelength of the laser and <math>\Delta\lambda</math> is the spectral width of the laser. The longer the coherence length, the further the photons can travel and still show interference effects. If the coherence length is shorter than the path difference between two beams of light, the interference pattern becomes less distinct. The calculated longest path for multiple scattered photons to achieve an interrogation depth in tissue of 25mm is around one meter. We require our laser to have longer than one meter coherence length. This property is crucial for our application of blood flow detection where we need to produce sharp, well-defined interference patterns. In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial mode onto the beam profiling camera. == Chirp == [[File:Chirp effect.png|thumb|''Figure 4 the interferometer signals with and without chirp (left) and the corresponding speckle images (right)''|412x412px]] Laser chirp refers to the phenomenon where the frequency or wavelength of a laser varies over time. It's a concept commonly encountered in ultrafast lasers or in lasers used for applications like optical communications, spectroscopy, and high-speed optical signal processing. In lasers, chirping and mode hopping can occur mainly due to : * Change in the refractive index of the gain medium due to the sudden change in current or charge carrier density. * Heating Effects: when a high and fast current flows through a laser, some amount of heat is produced. This disrupts the thermal equilibrium and affects the laser cavity and results in the chirping and mode-hopping. Managing and controlling laser chirp and mode hopping is essential in our measurement to ensure that our speckle signal is sufficient for repeatable and accurate measurement. We have developed techniques and devices to control chirping effects, allowing for a more stable laser output. This enhances measurement performance. Our tests are described below. The detrimental effect of the chirping is shown in Figure 4. The speckle contrast is significantly washed out due to the presence of the chirping during the laser pulse. Acceptable performance of Openwater blood flow monitoring systems, the laser must be highly coherent and frequency or wavelength stays stable during the laser pulse. ==Spatial Mode Test == ===Importance of spatial modes === In addition to the interferometric test, the laser is also tested for its transverse mode during pulsing by monitoring its spatial modes with a beam profiling camera. During the discovery phase laser testing for coherence length and blood flow measurements, it was found that laser systems differed from each other. The difference was traced to different amounts of transverse modes present in the amplifier section. This problem was solved by precise alignment of the seed laser to the amplifier. A measurement was developed to confirm alignment is optimized. === Laser spatial mode test procedure === [[File:Spatial test setup.jpg|thumb|251x251px|''Figure 5 Spatial mode test setup'']] Beam profiling is useful to investigate the spatial mode of the laser light. We look at the beam profile at the same time that we investigate its power level trace on an oscilloscope. This brief instruction explains how to perform beam profiling for the lasers used in Openwater systems. ==== Required test equipment ==== CMOS beam profiler Manufacturer: DataRay Part number: WinCamD-LCM ==== Experimental setup ==== Laser light exiting from the tapered amplifier (TA), passes through a free space optical isolator, and is directed towards the beam profiling camera through free space (see below). We run the TA with an operating current of about 5A in pulse mode with pulse width of ~ 100us, which corresponds to ~ 5W instantaneous power. Running the TA at lower currents is not recommended to investigate the beam profile, since higher currents result in multiple side lobes. We use ND filters to assure that the sensor is not saturated. High power beams can saturate the sensor distorting the measured beam profile and even damaging the camera sensor. To get a stable image on the camera, it should be synchronized with the pulsing laser. To do so, we sync the camera with the TA power supply. Here are the camera settings to use before capturing the beam profile: * Open the DataRay software * Trigger set up (Setup -> Setup Trigger): Choose TTL(Voltage)- positive edge * Exposure control (right click on “Imager Gain” or “Exposure time”): ** Enable auto gain adjustment ** Exposure time is not relevant due to external triggering * Make sure to zoom in to 3x for the center image Figures 5 and 6 show the beam profiles for 2 different lasers operating at 5A in pulse mode operation (TA current). [[File:Spetial test1.jpg|thumb|322x322px|''Figure 6 Good performance, single central lobe''|left]] [[File:Spatial test bad laser.jpg|thumb|332x332px|''Figure 7 Poor performance: multiple spatial modes'']] First beam profile was captured using a laser with good performance. As can be seen in Figure 6 there is one central mode with some side modes not carrying most of the laser energy. The 3D profile on the right side shows how much energy is distributed for each mode. A second beam profile was captured using a poor performing laser. Figure 7 shows that unlike the first one, the energy is distributed over several spatial modes and more importantly there are 2 central modes (lobes) in the center of the beam. The 3D profile clearly shows the energy distribution is far from ideal. This indicates that there could be mode hopping between the 2 center lobes. Since we couple all these modes to a multi-mode fiber, it will affect the speckle contrast measurement. The “side lobes” are due to the presence of the higher order transverse modes in the amplifier media and have higher frequencies than the central lobe. Due to the presence of these multiple higher frequency components in the beam the speckle contrast is reduced. The mode hopping between central lobes corresponds to the frequency of the laser hopping between multiple longitudinal modes during the pulse which also results in overriding noise in the measurement system and the reduction of the speckle contrast. == Interferometer Chirp Experiment == A significant challenge when pulsing our laser at high powers is keeping the wavelength stable during the pulse. We have observed that for many lasers, as soon as the pulse starts, the wavelength begins to increase (i.e. chirp). Unlike in continuous mode, in pulsed mode the laser does not have time to stabilize. As we show below, chirping of the lasers negatively impacts the laser coherence, and prevents us from measuring the speckle contrast. Our use of a MOPA configuration as described above, was motivated by the need to physically separate the laser cavity, which determines the wavelength, from the amplifier section. The master oscillator contains the laser cavity. It is operated continuously and maintains a stable wavelength. Pulsed operation is obtained by pulsing the current to the power amplifier, which does not affect the wavelength. Below we describe our experimental procedure for quantifying laser chirp using an interferometer. We compared the coherence and frequency and power stability by using the interferometric technique both qualitatively and quantitatively. This also serves the purpose of testing corrective schemes to mitigate the chirping and mode hopping issues during the pulse if needed. [[File:Michelson interferometer setup.png|thumb|512x512px|''Figure 8 A Michelson interferometer setup used for laser chirp measurements'']] === A description of the experiment is as follows: === ==== Introduction: ==== An interferometer has proven to be a very powerful diagnostic tool used in the diverse fields of science and engineering. We use a Michaelson interferometer to characterize the coherence behavior of our laser. In this interferometer, the laser from a single beam is split into two separate beams and recombined to study their coherence behavior by collecting the combined light either using a photo-detector or a camera. We use this behavior to judge the performance of different lasers for use in our system. ==== Michelson interferometer setup: ==== [[File:Hardwaredetails.png|thumb|512x512px|''Figure 9 the major components used to build the interferometer and the instruments used to drive PZT, monitor and collect the interferometer signals'' ]] The coherence is the capacity of the light to interfere. The coherence length becomes essential for the application that requires the speckle contrast measurements. Equivalently, narrower the linewidth of the lasers width, better will be the contrast. In order to primarily characterize the laser, the Michelson interferometer is devised as shown in the Figure 8. The hardware (optical elements, optomechanical and controller) used to build the simplified version of Michelson interferometer is shown in Figure 9. These are available in Thorlabs with the path number shown below each component. The output of the tapered amplifier (TA) after an optical isolator is split using (8/92) beam splitter (BS). The 8% reflected beam is aligned to the photodetector (PD1) to monitor the intensity of the laser and 92% of the beam is coupled into the interferometer. The beam is then further split into two beams with a second non-polarizing beam splitter cube 50/50 (BS). One of the beams is retro-reflected on TIR-retroreflector M1 at a shorter, fixed distance while the other is reflected at a longer, adjustable distance with a TIR-retroreflector M2 mounted onto the piezo-translation-stage. The delay is set to ~100 cm in the longer arm of the interferometer. The reflected beams from both arms are recombined. The recombined beam is then further split into two orthogonal beams via polarizing beam splitter cube (PBS) and are aligned to the photodetectors (PD2 and cam) as shown in the Figure 1 or replace cam with PD3 to obtain the phase information along with the coherence behavior of the laser. ==== Alignment and optimization of the interferometer: ==== [[File:InterferometerCamera1-2.gif|thumb|188x188px|''Figure 10b shows the fringes captured in the camera CAM during the ramp applied to the PZT'']] [[File:Ramp interferometer signal.png|left|thumb|258x258px|''In Figure 10a, a simple interferometer with the ramp applied to the piezoelectric transducer (PZT) attached to the mirror on the adjustable arm of the interferometer is shown in yellow, the blue signal is the interferometer signal collected in the PD2 in Figure 8'']] # Set the position of the adjustable arm of the interferometer. # Center the beams at the center of the mirrors M1 and M2 by adjusting two mirrors M and M after the laser. # Apply the ramp voltage to the PZT attached on the back of the mirror M2 as shown in the yellow signal in Figure 10a. # Optimize the interferometer signal by tweaking the mirror M2 to obtain an interferometer signal (blue) as shown in Figure 10a. # Alternatively, the camera can also be used to view the overlap of the two recombined beams. The Figure 10b is showing the fringe shifts corresponding to the ramp voltage applied to the PZT). ==== Example Data: ==== [[File:InterferometerData1.png|thumb|''Figure 11a and 11d show the interferometer signals collected at different adjustable arm mirror positions for good and bad lasers respectively. The Figure 11b and 11e are the intensity mean per pulse as a function of the mirror position and the Figure 11c and Figure 11f show the sample of phase variation during the laser pulse for single mirror position'']] The typical interferometer signals and the phasor diagrams are shown in Figure 11 for the comparison between good and underperforming TAs. The TA is driven at a 5A of current with pulse width of 100 us. A well performing TA shows smooth variation in the interference as we sweep the position of the adjustable arm mirror and the phase noise is tightly scattered compared to the underperforming TA. A bad TA shows broadly scattered phase noise and the interference pattern looks blurred. Figure 11b and 11e are the mean interferometer signal plotted against the mirror position. These show a significant reduction in the contrast for the underperforming laser shown in Figure 11e compared to that of the well-performing laser in Figure 11b. ==== Calculation of the path difference: ==== The calculation of more accurate path difference is required for the estimation of full fringe shift (FFS), maxima to nearest maxima in order to determine the accurate chirping during the pulsing of the laser. In order to determine the proper path difference we follow the procedure mentioned below. [[File:PathDifference 1.png|left|thumb|706x706px|''Figure 12 Schematic of the beam paths in the two arms of the interferometer with the position of the mirrors shown as M1 and M2. The exact positions of the mirrors are not known. Hence, we used the laser interference to measure the exact path difference'']] # Set the position of the adjustable arm-mirror mount with reference to the back edge of the base (xpos) as shown in Figure 12a. # Adjust and set the PZT voltage fixed to the maximum of the interferometer signal.[[File:PathdifferenceTable.png|thumb|484x484px|''Table1 wavelengths recorded at full fringe shift range at different mirror mount edge positions to estimate the reference position of the mirror at path difference equal to zero'']] # Note the wavelength of the laser on the wavelength meter (𝝀1). # Slowly and continuously change the temperature of the laser/heat-sink until the interferometer level drops to minimum and returns to the same maximum level. # Note the wavelength again on the wavelength meter (𝝀2). # Repeat the steps 1 - 5 for at least two more positions of the mirror. # The recorded wavelength and the mirror mount edge positions are shown in table 1. We define the path difference as [[File:FFS simulated.png|thumb|584x584px|''Figure 13a-13c show the simulated full fringe shift (FFS)) for the mirror mount edge positions of 420 mm, 370 mm and 320 mm respectively'']] path difference <math>= 2(x_2 - x_1)</math> Using the relation between the path difference, wavelength and the change the change in wavelength. We get the plot in Figure 12b. path difference <math>= \frac{\lambda^2}{\Delta\lambda}</math> A straight line is fitted and extrapolated. The x-intercept at path difference equal to zero provides the reference mirror mount edge position for our interferometer setup, <math> x_{pos0} </math> equals to ~ 452.35 mm. [[File:Chirp inteferometer12.png|thumb|583x583px|''Figure 14a and 14b show the interferometer signal (light blue), intensity (pink), driving current signal (dark blue) for the path difference of 6.3 cm and 26.3 cm respectively'']] path difference <math>= 2(452.35 - mirror\ mount\ edge\ position) \times 10^{-3}</math>m ==== Calculation of the full fringe shift range (FFS) for the setup: ==== Once the reference <math>x_{pos0}</math> is determined. We can calculate the FFS for any given path-differences. The FFS are simulated and shown in figures 13a-13c for the three mirror mount edge positions of 420, 370 and 320 respectively and outlined in column 5 table 1 to compare with the experimentally measured values in column 4. The measured interferometer signals for path difference of 6.5cm and 26.5 cm are shown in Figure 14a and 14b respectively. The smooth rate of change FFS indicates the non-linear change in junction temperature during the pulse which is shown in the figure 15b. [[File:Chirp plot12.png|thumb|584x584px|''Figure 15a shows the interferometer signal (light blue), smoothed signal (orange) and peak-position (yellow). Figure 15b is the plot of the chirping as a function of time'']] ==== Measurement of the chirping of the laser during pulse: ==== Finally, the mirror position is set to the base edge position x-pos equals to 40 mm equivalent to 82.47 cm of path difference. The interferometer signal is collected and plotted as shown in Figure 15a. The data is smoothed and determines the peak positions. Each successive peak separation corresponds to full fringe shift (FFS). The number of peaks are counted and the approximate total chirping during the laser pulse is estimated as shown in the Figure 15b. Some lasers can show mode-hopping during the pulse which can skew our chirping results but the approach still provides the test to qualify the laser system. ==== Measurement of junction temperature: ==== [[File:Junction temperature.png|thumb|581x581px|''Figure 16a wavelength versus waste power for different heat sink temperatures, Figure 16b shows relation between junction temperatures and the wavelength of the output laser'']] The chirping in the laser can be caused by a change in the driving current (carrier density) and the junction temperature. Since we are investigating the chirping 10 µs after the beginning of the driving pulse, the driving current already reaches the stable value. Therefore, all the chirping in the laser is caused due to the change in the junction temperature of the laser. In order to measure the junction temperature, the LIV and wavelength data are collected for a set of fixed heat sink temperatures and a set of driving currents above threshold ensuring the thermal equilibrium has reached before each data collection [11.1]. The waste power is calculated from the data using the relation <math>P_{j} = I \times V - L </math> where <math> P_{j} </math> is the waste power, I V and L are driving current, voltage and the output laser power respectively. The corresponding measured wavelength is plotted against the calculated waste power. The data is fitted to a line and extrapolated, shown in Figure 9a. The intercept when the waste power is zero corresponds to the wavelength when the junction temperature Tj is equal to the heat sink temperature . Now, using the wavelengths at the intercept for different heat sink temperatures a relation between them can be established as shown in Figure 16b. The slope of the fitted line in figure 16b gives the wavelength dependence on the temperature. This is equal to <math>0.084nm/^oC</math>. Using the relation between Tj and wavelength of the laser <math>\Delta T_{j} = 11.49 \times \Delta \lambda = 2.56 ^oC</math>. ==== Conclusion: ==== The laser under test shows the chirping of <math> \Delta \lambda \approx 0.223 nm </math> corresponding to <math> \approx 77 GHz</math>. This drift in the wavelength or frequency during the laser pulse corresponds to the change in temperature by <math>\approx 2.6^oC</math>. The amount of acceptable chirping is <math> \leq 65 MHz</math>. Therefore, the laser shown in figure 15 is not acceptable for the speckle contrast measurement application due to the presence of the chirp washing out the speckle contrast. We require improved heat dissipation in our laser system with the thermal stability to the level of <math> \Delta T_{j} \leq 2E-3 ^oC</math> within the pulse width <math> \geq 250 \mu s</math>. == Speckle Contrast Measurements == [[File:XIMEA111.gif|thumb|232x232px|Figure 17 speckle pattern ]] After characterization of the laser system in the spatial mode test and interferometric chirp test, the laser is run through the direct speckle contrast measurement test. This test measures the contrast on the speckle pattern formed on the camera. In general, longer coherence length (narrower linewidth) results in higher speckle contrast, which in turn improves our ability to detect blood flow. The setup and the step-by-step procedure are outlined below: === Introduction: === Openwater’s blood flow measurement is based on the analysis of the remitted light from human tissues on a camera sensor. When the coherent light is detected on our camera sensors, it produces a speckle pattern as shown in the figure 17. We measure blood flow by quantifying the reduction in contrast that occurs when the detected light has scattered off moving objects such as red blood cells. Thus it is of critical importance that our lasers and cameras produce stable images with high contrast, so that we can measure the decrease in contrast due to blood flow. [[File:Speckle test setup.png|thumb|395x395px|''Figure 19a and 19b depict the setup for the speckle contrast measurements. The laser beam enters the integrating cubic integrating sphere. The multiple scattered beams make it through the 700 um pinhole ~90mm away from the XIMEA camera and form a speckle as shown in the figure 17'']] === Hardware details: === The essential hardware used to build the speckle contrast measurement setup are shown in the figure 18. [[File:Hardware speckle test.png|none|thumb|763x763px|''Figure 18 Hardware components used in speckle measurement setup'']] === Speckle measurement setup: === The laser beam out of the tapered amplifier (TA) and optical isolator is coupled to the integrating sphere through an input port. The integrating sphere is attached to the camera through the pinhole and the lens tube on the side orthogonal to the input port as shown in the figure 19a-19b. The coupled beam undergoes multiple scattering inside the integrating sphere and passes through the 700 µm pinhole and forms a speckle pattern on the camera mounted 9 cm away from the pinhole. === XIMEA camera set-up and software settings: === # [[File:Speckle setup trigger.jpg|thumb|154x154px|''Figure 20 shows trigger source to sync the Ximea camera and laser pulse'']]Connect the camera's USB cable to the USB3 connection on the laptop or… # To synchronize the camera and the laser pulse, connect the IO-USB-Power port on the back of the camera to the trigger output of the laser driver (Thorlabs ITC4020) as shown in figure 20: ## The Ximea’s trigger needs to be delayed to properly not clip the 100-250 us pulse. Therefore a delay generator (with the settings in Figure 21, channel 1) will need to be used. Connect the “TRIGGER OUT TTL 5V'' output on the back of the Thorlabs ITC4020 to the “CH1: Mod/TSK/Trig” on the delay generator and the Output of Channel 1 should be connected to the camera’s trigger connection. # Open the “XIMEA CamTool” software. # [[File:XimeaTrig.png|thumb|''Figure 21 ch1 settings for triggering the XIMEA camera'']]Figure 22 shows the software parameters (when connected) as explained below: # Software settings: Before running the software, complete the following settings and then start image acquisition. ## From the “Tools” menu at the top, make sure that the following features are selected: LUTs, Camera Frame rate, Histogram (statistics) ## Basic Settings: ### No downsampling, choose 1(4112x3008) option ### From the next dropdown menu select “Raw16” option ### Do NOT select “Auto exposure” option ### Set the exposure time to 10ms ### On Gain selector, select Analog ### Set Gain at 0dB ### ROI set on, defined as following (full-frame will not work properly at 30 Hz): #### Offset X: 1032 #### Offset Y: 752 #### Width: 2056 #### Height: 1504 ## Performance: ### On Control FPS select “free run” option ## Trigger, Device I/O ### Mode: rising edge ### Device I/O set up: On input for GPI1 select trigger ### Selector: exposure active ### Burst count: 0 ## LUTs: ### Select the “Auto” box ## Other: ### Set the gamma settings of the camera are set to 1.0 to remove any alteration of the raw values. At this point start acquisition by pressing the run button at the top left of the software and make sure that camera frame rate is the same as laser repetition rate. If they are not the same, repeat “Trigger, Device I/O” settings one more time and also manually change the exposure time and set it back to 10ms again. # [[File:Speckle and dark.jpg|thumb|''Figure 22 shows the tabs for dark count and signal along with the numerical count display on the bottom right of the figure'']]Dark count and signal: ## After the measurement is completed, find a couple of tabs on the software (Figure 22). Browse through the different tabs to find the dark count and signal. These counts can be found on the bottom of the software as labeled dark and signal. The tab titled “Image from script” contains the dark count and the one titled “MC124MG…” contains detected signals (dark included). # Recording speckle contrast measurement: ## [[File:SpeckleScript result.png|thumb|287x287px|''Figure 23 Scripts and results display in the output console'']]Using a script: XIMEA_Specke_Contrast_7.lua in repository - software measures many images and records the speckle contrast for all images. To do so, from the “Plugins” menu on the top, select “Script Editor”. This opens the script (Figure 23). Number of images can be set to as many as necessary on the script. The default is 500 images. This script also measures the background noise. ### https://www.ximea.com/support/wiki/apis/xiapi_manual ## First make sure both the tapered amplifier and seed laser are turned off. ### NOTE, a beam block is much preferred to turning both the seed and TA on and off between every measurement, which can introduce variability depending on how long the seed and TA are left on to stabilize between each measurement. ## Next press the Run button on the script. This will measure the background noise ## Now turn the seed laser and tapered amplifier and then press ok button to measure the speckle contrast. ##After the measurement is completed, click on the “Output” section on the lower part of the script and select all (shown below). Now all the data can be copied to an excel sheet. ## Select the column with the data on the excel sheet and from the data tab of the excel sheet select “Text to columns” option and choose “comma”, “space” and “Tab delimited” and “finish”. ## Now all the data is stored in columns for the 500 images measured. The last column is the speckle contrast, which is calculated by dividing the adjacent columns of standard deviation over average. ##The mean and standard deviation of the 500 speckle contrast measurements can be calculated. == References == #“Measuring high power laser diode junction temperature and package thermal impedance, Lawrence A. Johnson and Andrew Teh” 9e92uvx0ocmyb9axissnj0rl7kks2kh 0 76 343 2023-12-14T00:43:25Z Admin 1 Admin moved page [[Main Page]] to [[Openwater Wiki]] 343 wikitext text/x-wiki #REDIRECT [[Openwater Wiki]] nbpkj7t6049lmlr3ckibk0r74ei1ih0 0 55 178 2023-12-13T16:30:17Z OpenwaterPeterH 9 Neuromodulation Platform 178 wikitext text/x-wiki # Open-TFUS Neuromodulation Platform ## Overview Open-TFUS is an ultrasound platform designed to help qualified researchers transmit focused ultrasound beams into subject’s brains, so that those researchers can learn more about how different types of ultrasound beams interact with the neurons in the brain. Unlike other focused ultrasound systems which are aimed only by their placement on the head, Open-TFUS uses an array to precisely steer the ultrasound focus to the target location, while its small size allows transmission through the forehead.  The first application researched using this platform has been Neuromodulation, which is when brain activity in a specific part of the brain is enhanced or suppressed by the application of energy - in this case, ultrasound energy. Neuromodulation is unique in that the therapeutic effect can be achieved with Low Intensity Focused Ultrasound (LOFU), meaning that the energy levels needed to create the therapeutic effect are below the limits understood to be safe for diagnostic imaging. The exact level used, the timing of pulses, and where the beam(s) are aimed are up to the researcher and specific application, but Open-TFUS supports a wide range of acoustic parameters to support Neuromodulation and other focused ultrasound applications research. A system built on open-TFUS is made of: * A headset containing an ultrasound transducer that is aimed through the forehead * A system for recording the precise position of the transducer relative to the subject’s MRI * Software for computing what signals the transducer will have to send to treat the target * Hardware for driving the transducer to generate those signals and deliver the treatment. 2ol4ayacncp92g28ajzeqy5f3li4cmc 180 178 2023-12-13T16:35:08Z OpenwaterPeterH 9 Added Overview 180 wikitext text/x-wiki == Open-TFUS Neuromodulation Platform == === Overview === Open-TFUS is an ultrasound platform designed to help qualified researchers transmit focused ultrasound beams into subject’s brains, so that those researchers can learn more about how different types of ultrasound beams interact with the neurons in the brain. Unlike other focused ultrasound systems which are aimed only by their placement on the head, Open-TFUS uses an array to precisely steer the ultrasound focus to the target location, while its small size allows transmission through the forehead. [[File:289990844-1e7a97e3-3d13-4f30-9f0e-1487edf5eeaf.png|thumb|Drawing of Neuromodulation System]] The first application researched using this platform has been Neuromodulation, which is when brain activity in a specific part of the brain is enhanced or suppressed by the application of energy - in this case, ultrasound energy. Neuromodulation is unique in that the therapeutic effect can be achieved with Low Intensity Focused Ultrasound (LOFU), meaning that the energy levels needed to create the therapeutic effect are below the limits understood to be safe for diagnostic imaging. The exact level used, the timing of pulses, and where the beam(s) are aimed are up to the researcher and specific application, but Open-TFUS supports a wide range of acoustic parameters to support Neuromodulation and other focused ultrasound applications research. A system built on open-TFUS is made of: * A headset containing an ultrasound transducer that is aimed through the forehead * A system for recording the precise position of the transducer relative to the subject’s MRI * Software for computing what signals the transducer will have to send to treat the target * Hardware for driving the transducer to generate those signals and deliver the treatment. === Getting Started === ==== Required Software ==== 1. Download Matlab (2021a) and license 2. Download [K-Wave](<nowiki>http://www.k-wave.org/download.php</nowiki>) 3. Download the [K-Wave alpha functions](<nowiki>http://www.k-wave.org/downloads/kWaveArray_alpha_0.3.zip</nowiki>) and add them to the K-Wave installation folder (specifically we need `kWaveArray.m`) 4. If using Verasonics, install Verasonics software, versions 4.6.2, and request a license. ==== Examples ==== Example scripts can be found in the [https://github.com/OpenwaterInternet/opw_neuromod_sw/tree/main/examples examples] folder. These cover a number of topics, including the initial creation of a database and specification of an ultrasound transducer, as well as the planning and delivery of a treatment. ==== User Manual ==== For a detailed walkthrough of using the GUI, see the [https://github.com/OpenwaterHealth/opw_neuromod_sw/wiki/1.-User-Manual User Manual] fbqwjlifiwfdp10oqrfa78tb9q3hww1 186 180 2023-12-13T16:39:27Z OpenwaterPeterH 9 add links to repos 186 wikitext text/x-wiki == Open-TFUS Neuromodulation Platform == === Overview === Open-TFUS is an ultrasound platform designed to help qualified researchers transmit focused ultrasound beams into subject’s brains, so that those researchers can learn more about how different types of ultrasound beams interact with the neurons in the brain. Unlike other focused ultrasound systems which are aimed only by their placement on the head, Open-TFUS uses an array to precisely steer the ultrasound focus to the target location, while its small size allows transmission through the forehead. [[File:289990844-1e7a97e3-3d13-4f30-9f0e-1487edf5eeaf.png|thumb|Drawing of Neuromodulation System]] The first application researched using this platform has been Neuromodulation, which is when brain activity in a specific part of the brain is enhanced or suppressed by the application of energy - in this case, ultrasound energy. Neuromodulation is unique in that the therapeutic effect can be achieved with Low Intensity Focused Ultrasound (LOFU), meaning that the energy levels needed to create the therapeutic effect are below the limits understood to be safe for diagnostic imaging. The exact level used, the timing of pulses, and where the beam(s) are aimed are up to the researcher and specific application, but Open-TFUS supports a wide range of acoustic parameters to support Neuromodulation and other focused ultrasound applications research. A system built on open-TFUS is made of: * A headset containing an ultrasound transducer that is aimed through the forehead * A system for recording the precise position of the transducer relative to the subject’s MRI * Software for computing what signals the transducer will have to send to treat the target * Hardware for driving the transducer to generate those signals and deliver the treatment. === Getting Started === ==== Repositories ==== * [https://github.com/OpenwaterHealth/opw_neuromod_hw opw_neuromod_hw]: instruction manuals, mechanical drawings and 3D models * [https://github.com/OpenwaterHealth/opw&#x20;neuromod&#x20;sw opw_neuromod_sw]: software for configuring and controlling systems ==== Required Software ==== 1. Download Matlab (2021a) and license 2. Download [K-Wave](<nowiki>http://www.k-wave.org/download.php</nowiki>) 3. Download the [K-Wave alpha functions](<nowiki>http://www.k-wave.org/downloads/kWaveArray_alpha_0.3.zip</nowiki>) and add them to the K-Wave installation folder (specifically we need `kWaveArray.m`) 4. If using Verasonics, install Verasonics software, versions 4.6.2, and request a license. ==== Examples ==== Example scripts can be found in the [https://github.com/OpenwaterInternet/opw_neuromod_sw/tree/main/examples examples] folder. These cover a number of topics, including the initial creation of a database and specification of an ultrasound transducer, as well as the planning and delivery of a treatment. ==== User Manual ==== For a detailed walkthrough of using the GUI, see the [https://github.com/OpenwaterHealth/opw_neuromod_sw/wiki/1.-User-Manual User Manual] pqu3z44ncrgpsel0jeda0a2jkgedq8q 296 186 2023-12-13T22:12:31Z OpenwaterPeterH 9 Remove info mirrored in repo 296 wikitext text/x-wiki == Open-TFUS Neuromodulation Platform == === Overview === Open-TFUS is an ultrasound platform designed to help qualified researchers transmit focused ultrasound beams into subject’s brains, so that those researchers can learn more about how different types of ultrasound beams interact with the neurons in the brain. Unlike other focused ultrasound systems which are aimed only by their placement on the head, Open-TFUS uses an array to precisely steer the ultrasound focus to the target location, while its small size allows transmission through the forehead. [[File:289990844-1e7a97e3-3d13-4f30-9f0e-1487edf5eeaf.png|thumb|Drawing of Neuromodulation System]] The first application researched using this platform has been Neuromodulation, which is when brain activity in a specific part of the brain is enhanced or suppressed by the application of energy - in this case, ultrasound energy. Neuromodulation is unique in that the therapeutic effect can be achieved with Low Intensity Focused Ultrasound (LIFU), meaning that the energy levels needed to create the therapeutic effect are below the limits understood to be safe for diagnostic imaging. The exact level used, the timing of pulses, and where the beam(s) are aimed are up to the researcher and specific application, but Open-TFUS supports a wide range of acoustic parameters to support Neuromodulation and other focused ultrasound applications research. A system built on open-TFUS is made of: * A headset containing an ultrasound transducer that is aimed through the forehead * A system for recording the precise position of the transducer relative to the subject’s MRI * Software for computing what signals the transducer will have to send to treat the target * Hardware for driving the transducer to generate those signals and deliver the treatment. === Getting Started === ==== Repositories ==== * [https://github.com/OpenwaterHealth/opw_neuromod_hw opw_neuromod_hw]: instruction manuals, mechanical drawings and 3D models * [https://github.com/OpenwaterHealth/opw&#x20;neuromod&#x20;sw opw_neuromod_sw]: software for configuring and controlling systems ==== User Manual ==== For a detailed walkthrough of using the GUI, see the [https://github.com/OpenwaterHealth/opw_neuromod_sw/wiki/1.-User-Manual User Manual] d27cq9fczg175hptwc44ayu8lzp3v9j 373 296 2023-12-14T18:07:43Z 75.182.93.123 changed link 373 wikitext text/x-wiki == Open-TFUS Neuromodulation Platform == === Overview === Open-TFUS is an ultrasound platform designed to help qualified researchers transmit focused ultrasound beams into subject’s brains, so that those researchers can learn more about how different types of ultrasound beams interact with the neurons in the brain. Unlike other focused ultrasound systems which are aimed only by their placement on the head, Open-TFUS uses an array to precisely steer the ultrasound focus to the target location, while its small size allows transmission through the forehead. [[File:289990844-1e7a97e3-3d13-4f30-9f0e-1487edf5eeaf.png|thumb|Drawing of Neuromodulation System]] The first application researched using this platform has been Neuromodulation, which is when brain activity in a specific part of the brain is enhanced or suppressed by the application of energy - in this case, ultrasound energy. Neuromodulation is unique in that the therapeutic effect can be achieved with Low Intensity Focused Ultrasound (LIFU), meaning that the energy levels needed to create the therapeutic effect are below the limits understood to be safe for diagnostic imaging. The exact level used, the timing of pulses, and where the beam(s) are aimed are up to the researcher and specific application, but Open-TFUS supports a wide range of acoustic parameters to support Neuromodulation and other focused ultrasound applications research. A system built on open-TFUS is made of: * A headset containing an ultrasound transducer that is aimed through the forehead * A system for recording the precise position of the transducer relative to the subject’s MRI * Software for computing what signals the transducer will have to send to treat the target * Hardware for driving the transducer to generate those signals and deliver the treatment. === Getting Started === ==== Repositories ==== * [https://github.com/OpenwaterHealth/opw_neuromod_hw opw_neuromod_hw]: instruction manuals, mechanical drawings and 3D models * [https://github.com/OpenwaterHealth/opw_neuromod_sw opw_neuromod_sw]: software for configuring and controlling systems ==== User Manual ==== For a detailed walkthrough of using the GUI, see the [https://github.com/OpenwaterHealth/opw_neuromod_sw/wiki/1.-User-Manual User Manual] 457o8zglrtss34jxv4kmzyl6b5bo3yp 631 373 2023-12-19T18:42:28Z Admin 1 Protected "[[Neuromodulation]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 373 wikitext text/x-wiki == Open-TFUS Neuromodulation Platform == === Overview === Open-TFUS is an ultrasound platform designed to help qualified researchers transmit focused ultrasound beams into subject’s brains, so that those researchers can learn more about how different types of ultrasound beams interact with the neurons in the brain. Unlike other focused ultrasound systems which are aimed only by their placement on the head, Open-TFUS uses an array to precisely steer the ultrasound focus to the target location, while its small size allows transmission through the forehead. [[File:289990844-1e7a97e3-3d13-4f30-9f0e-1487edf5eeaf.png|thumb|Drawing of Neuromodulation System]] The first application researched using this platform has been Neuromodulation, which is when brain activity in a specific part of the brain is enhanced or suppressed by the application of energy - in this case, ultrasound energy. Neuromodulation is unique in that the therapeutic effect can be achieved with Low Intensity Focused Ultrasound (LIFU), meaning that the energy levels needed to create the therapeutic effect are below the limits understood to be safe for diagnostic imaging. The exact level used, the timing of pulses, and where the beam(s) are aimed are up to the researcher and specific application, but Open-TFUS supports a wide range of acoustic parameters to support Neuromodulation and other focused ultrasound applications research. A system built on open-TFUS is made of: * A headset containing an ultrasound transducer that is aimed through the forehead * A system for recording the precise position of the transducer relative to the subject’s MRI * Software for computing what signals the transducer will have to send to treat the target * Hardware for driving the transducer to generate those signals and deliver the treatment. === Getting Started === ==== Repositories ==== * [https://github.com/OpenwaterHealth/opw_neuromod_hw opw_neuromod_hw]: instruction manuals, mechanical drawings and 3D models * [https://github.com/OpenwaterHealth/opw_neuromod_sw opw_neuromod_sw]: software for configuring and controlling systems ==== User Manual ==== For a detailed walkthrough of using the GUI, see the [https://github.com/OpenwaterHealth/opw_neuromod_sw/wiki/1.-User-Manual User Manual] 457o8zglrtss34jxv4kmzyl6b5bo3yp 662 631 2023-12-21T17:09:35Z OpenwaterPeterH 9 Updated info 662 wikitext text/x-wiki == A New Paradigm in Focused Ultrasound == Open-LIFU is the first commercial open-source platform for planning and delivering Low Intensity Focused Ultrasound (LIFU) sequences. For researchers investigating novel applications of transcranial focused ultrasound (tFUS), Open-LIFU enables precise steering of ultrasound beams to targets up to 6 cm deep beneath the forehead. Unlike conventional tFUS fixed-focus transducers which must be precisely placed by hand on the head to exactly align their deep focus with the target inside the brain, Open-LIFU uses a “matrix-in-headset” design that allows for much more rapid setup because the transducer only needs to be placed with the target “in view”, using the precise steering from the array to automatically fine-tune the focus to the target. This steerable design also supports advanced beamforming techniques, including compensation for skull geometry and rastering of the focal spot across multiple locations within treatment sequences, while maintaining subject comfort with a lightweight and wearable headset. [[File:289990844-1e7a97e3-3d13-4f30-9f0e-1487edf5eeaf.png|thumb|Drawing of Neuromodulation System]] = System Overview = A system built on Open-LIFU is made of: * A headset containing an ultrasound transducer that is aimed through the forehead * A neuronavigation system for recording the precise position of the transducer relative to the subject’s MRI * Software for computing what signals the transducer will have to send to treat the target * Hardware for driving the transducer to generate those signals and deliver the treatment. [[File:OpenLIFUBlockDiagram.png|thumb|Open-LIFU Block Diagram]] = Development = While the initial iteration of Open-LIFU utilizes 3rd-party hardware for neuronavigation (Localite TMSNavigator) and treatment delivery (Verasonics Vantage 128), Openwater is working on its next generation version that will use optimized and open-source hardware that is smaller, cheaper, and suitable for partners wishing to commercialize open-source medical devices on the platform. The next version will also migrate the required MATLAB interface to Python and support a modular design for beamforming across multiple arrays for focusing even deeper in the brain with a wide range of frequencies. However, the principle of operation and data models are designed to be compatible across generations, allowing research discoveries made on the existing platform to seamlessly integrate into the next version. = Applications = In its initial configuration, the platform has been used to explore tFUS-induced neuromodulation in the brains of patients with major depressive disorder.  Ultrasound Neuromodulation is when brain activity in a specific part of the brain is enhanced or suppressed by the application of focused ultrasound, which researchers believe may be therapeutic for subjects with certain neurological conditions. Because of the steerable and configurable nature of the Open-LIFU platform, is also well-suited to adaptation for other neuromodulation tasks, and tFUS tasks more generally. == Getting Started == === Repositories === * [https://github.com/OpenwaterHealth/opw_neuromod_hw opw_neuromod_hw]: instruction manuals, mechanical drawings and 3D models * [https://github.com/OpenwaterHealth/opw_neuromod_sw opw_neuromod_sw]: software for configuring and controlling systems === User Manual === For a detailed walkthrough of using the GUI, see the [https://github.com/OpenwaterHealth/opw_neuromod_sw/wiki/1.-User-Manual User Manual] ''CAUTION - Investigational device. Limited by Federal (or United States) law to investigational use. Open-LIFU has not been evaluated by the FDA and is not designed for the treatment or diagnosis of any disease. It is provided AS-IS, with no warranties. User assumes all liability and responsibility for identifying and mitigating risks associated with use'' bs633c546h7wft0dvgn8elbqjoguds0 665 662 2023-12-22T19:45:59Z KedarGrama 6 Reverted edits by [[Special:Contributions/OpenwaterPeterH|OpenwaterPeterH]] ([[User talk:OpenwaterPeterH|talk]]) to last revision by [[User:Admin|Admin]] 373 wikitext text/x-wiki == Open-TFUS Neuromodulation Platform == === Overview === Open-TFUS is an ultrasound platform designed to help qualified researchers transmit focused ultrasound beams into subject’s brains, so that those researchers can learn more about how different types of ultrasound beams interact with the neurons in the brain. Unlike other focused ultrasound systems which are aimed only by their placement on the head, Open-TFUS uses an array to precisely steer the ultrasound focus to the target location, while its small size allows transmission through the forehead. [[File:289990844-1e7a97e3-3d13-4f30-9f0e-1487edf5eeaf.png|thumb|Drawing of Neuromodulation System]] The first application researched using this platform has been Neuromodulation, which is when brain activity in a specific part of the brain is enhanced or suppressed by the application of energy - in this case, ultrasound energy. Neuromodulation is unique in that the therapeutic effect can be achieved with Low Intensity Focused Ultrasound (LIFU), meaning that the energy levels needed to create the therapeutic effect are below the limits understood to be safe for diagnostic imaging. The exact level used, the timing of pulses, and where the beam(s) are aimed are up to the researcher and specific application, but Open-TFUS supports a wide range of acoustic parameters to support Neuromodulation and other focused ultrasound applications research. A system built on open-TFUS is made of: * A headset containing an ultrasound transducer that is aimed through the forehead * A system for recording the precise position of the transducer relative to the subject’s MRI * Software for computing what signals the transducer will have to send to treat the target * Hardware for driving the transducer to generate those signals and deliver the treatment. === Getting Started === ==== Repositories ==== * [https://github.com/OpenwaterHealth/opw_neuromod_hw opw_neuromod_hw]: instruction manuals, mechanical drawings and 3D models * [https://github.com/OpenwaterHealth/opw_neuromod_sw opw_neuromod_sw]: software for configuring and controlling systems ==== User Manual ==== For a detailed walkthrough of using the GUI, see the [https://github.com/OpenwaterHealth/opw_neuromod_sw/wiki/1.-User-Manual User Manual] 457o8zglrtss34jxv4kmzyl6b5bo3yp 667 665 2024-01-02T16:13:56Z OpenwaterPeterH 9 Re-update to latest language 667 wikitext text/x-wiki === Open-LIFU Neuromodulation Platform - A New Paradigm in Focused Ultrasound === [[File:289990844-1e7a97e3-3d13-4f30-9f0e-1487edf5eeaf.png|thumb|Open-LIFU]] Open-LIFU is the first commercial open-source platform for planning and delivering Low Intensity Focused Ultrasound (LIFU) sequences. For researchers investigating novel applications of transcranial focused ultrasound (tFUS), Open-LIFU enables precise steering of ultrasound beams to targets up to 6 cm deep beneath the forehead. Unlike conventional tFUS fixed-focus transducers which must be precisely placed by hand on the head to exactly align their deep focus with the target inside the brain, Open-LIFU uses a “matrix-in-headset” design that allows for much more rapid setup because the transducer only needs to be placed with the target “in view”, using the precise steering from the array to automatically fine-tune the focus to the target. This steerable design also supports advanced beamforming techniques, including compensation for skull geometry and rastering of the focal spot across multiple locations within treatment sequences, while maintaining subject comfort with a lightweight and wearable headset. === System Overview === [[File:OpenLIFUBlockDiagram.png|thumb|System Block Diagram]] A system built on Open-LIFU is made of: * A headset containing an ultrasound transducer that is aimed through the forehead * A neuronavigation system for recording the precise position of the transducer relative to the subject’s MRI * Software for computing what signals the transducer will have to send to treat the target * Hardware for driving the transducer to generate those signals and deliver the treatment. === Development === While the initial iteration of Open-LIFU utilizes 3rd-party hardware for neuronavigation (Localite'''™''' TMSNavigator'''™''') and treatment delivery (Verasonics'''™''' Vantage 128'''™'''), Openwater is working on its next generation version that will use optimized and open-source hardware that is smaller, cheaper, and suitable for partners wishing to commercialize open-source medical devices on the platform. The next version will also migrate the required MATLAB interface to Python and support a modular design for beamforming across multiple arrays for focusing even deeper in the brain with a wide range of frequencies. However, the principle of operation and data models are designed to be compatible across generations, allowing research discoveries made on the existing platform to seamlessly integrate into the next version. === Applications === In its initial configuration, the platform has been used to explore tFUS-induced neuromodulation in the brains of patients with major depressive disorder.  Ultrasound Neuromodulation is when brain activity in a specific part of the brain is enhanced or suppressed by the application of focused ultrasound, which researchers believe may be therapeutic for subjects with certain neurological conditions. Because of the steerable and configurable nature of the Open-LIFU platform, is also well-suited to adaptation for other neuromodulation tasks, and tFUS tasks more generally. === Getting Started === To get started setting up a system with Open-LIFU, you can refer to the user manual and documentation in the following repositories: * [https://github.com/OpenwaterHealth/opw_neuromod_hw opw_neuromod_hw]: instruction manuals, mechanical drawings and 3D models * [https://github.com/OpenwaterHealth/opw_neuromod_sw opw_neuromod_sw]: software for configuring and controlling systems ==== User Manual ==== For a detailed walkthrough of using the GUI, see the [https://github.com/OpenwaterHealth/opw_neuromod_sw/wiki/1.-User-Manual User Manual] qzw9szmfpmj9zqcra72tnoqdm5cpxkf 669 667 2024-01-02T23:24:46Z KedarGrama 6 669 wikitext text/x-wiki === Open-LIFU Neuromodulation Platform - A New Paradigm in Focused Ultrasound === [[File:289990844-1e7a97e3-3d13-4f30-9f0e-1487edf5eeaf.png|thumb|Open-LIFU]] Open-LIFU is the first commercial open-source platform for planning and delivering Low Intensity Focused Ultrasound (LIFU) sequences. For researchers investigating novel applications of transcranial focused ultrasound (tFUS), Open-LIFU enables precise steering of ultrasound beams to targets up to 6 cm deep beneath the forehead. Unlike conventional tFUS fixed-focus transducers which must be precisely placed by hand on the head to exactly align their deep focus with the target inside the brain, Open-LIFU uses a “matrix-in-headset” design that allows for much more rapid setup because the transducer only needs to be placed with the target “in view”, using the precise steering from the array to automatically fine-tune the focus to the target. This steerable design also supports advanced beamforming techniques, including compensation for skull geometry and rastering of the focal spot across multiple locations within treatment sequences, while maintaining subject comfort with a lightweight and wearable headset. === System Overview === [[File:OpenLIFUBlockDiagram.png|thumb|System Block Diagram]] A system built on Open-LIFU is made of: * A headset containing an ultrasound transducer that is aimed through the forehead * A neuronavigation system for recording the precise position of the transducer relative to the subject’s MRI * Software for computing what signals the transducer will have to send to treat the target * Hardware for driving the transducer to generate those signals and deliver the treatment. === Development === While the initial iteration of Open-LIFU utilizes 3rd-party hardware for neuronavigation (Localite'''™''' TMSNavigator'''™''') and treatment delivery (Verasonics'''™''' Vantage 128'''™'''), Openwater is working on its next generation version that will use optimized and open-source hardware that is smaller, cheaper, and suitable for partners wishing to commercialize open-source medical devices on the platform. The next version will also migrate the required MATLAB interface to Python and support a modular design for beamforming across multiple arrays for focusing even deeper in the brain with a wide range of frequencies. However, the principle of operation and data models are designed to be compatible across generations, allowing research discoveries made on the existing platform to seamlessly integrate into the next version. === Applications === In its initial configuration, the platform has been used to explore tFUS-induced neuromodulation in the brains of patients with major depressive disorder.  Ultrasound Neuromodulation is when brain activity in a specific part of the brain is enhanced or suppressed by the application of focused ultrasound, which researchers believe may be therapeutic for subjects with certain neurological conditions. Because of the steerable and configurable nature of the Open-LIFU platform, is also well-suited to adaptation for other neuromodulation tasks, and tFUS tasks more generally. === Getting Started === To get started setting up a system with Open-LIFU, you can refer to the user manual and documentation in the following repositories: * [https://github.com/OpenwaterHealth/opw_neuromod_hw opw_neuromod_hw]: instruction manuals, mechanical drawings and 3D models * [https://github.com/OpenwaterHealth/opw_neuromod_sw opw_neuromod_sw]: software for configuring and controlling systems ==== User Manual ==== For a detailed walkthrough of using the GUI, see the [https://github.com/OpenwaterHealth/opw_neuromod_sw/wiki/1.-User-Manual User Manual] ==== Publications ==== [https://www.medrxiv.org/content/10.1101/2023.12.22.23300243v1 Publication] on the clinical study with our first generation system. m4ausg8w7kmw9mzomt9utdrq0dl4poq 0 53 174 2023-12-13T15:31:35Z OpenwaterPeterH 9 Oncolysis System 174 wikitext text/x-wiki # Preclinical Oncolysis Welcome to the opw_oncolysis_sw wiki! This describes the functionality of the Open-TFUS software found at https://github.com/OpenwaterInternet/opw_oncolysis_sw # Getting Started ## Installation Instructions Install/Run 32-bit Python 3.10+ (MUST BE 32-bit): `<path to 32 bit python>\python.exe install.py` This creates a virtual environment called env with the required dependencies. On Windows 10, Python may be installed to `C:\Users\<uname>\AppData\Local\Programs\Python\Python310-32` Install IVI Compliance Package 21.0 https://www.ni.com/en-us/support/downloads/drivers/download/packaged.ivi-compliance-package.409836.html Install DG4000 IVI Driver https://www.rigolna.com/products/waveform-generators/dg4000/ Install UltraSigma Instrument Connectivity Driver https://beyondmeasure.rigoltech.com/acton/attachment/1579/u-0003/0/-/-/-/-/ Enable Windows to load drivers from unknown sources: https://www.isunshare.com/windows-11/how-to-disable-driver-signature-enforcement-on-windows-11.html Download and install the Radiall USB Drivers from https://www.radiall.com/products/rf-microwave-switches/usb-coaxial-switches.html. Copy the folder containing `CP210xRuntime.dll`, `Radial_USBInterface.dll` and `Radial_USBInterface.xml` into a folder called `dll` in the root directory of the project. # License This Project is licensed under the GNU Affero General Public License v3.0. See [LICENSE](https://github.com/OpenwaterInternet/opw_oncolysis_sw/blob/main/LICENSE) for details. # Contributing See [Contributor Guidelines](Contributor%20Guidelines) for details. [Report an Issue](https://goo.gl/forms/chVYUnA4bP70WGsL2) duzxbzn7tk4mkl0yzs78ha7jq8jbfvt 176 174 2023-12-13T15:42:26Z OpenwaterPeterH 9 Uploaded summary 176 wikitext text/x-wiki == Preclinical Oncolysis System == === Overview === The Openwater Preclinical Oncolysis Prototype is designed to help researchers investigate the effect of different ultrasound parameters on a variety of in vitro and preclinical in vivo targets. Certain acoustic parameters may be well suited for damaging cancer cells while sparing surrounding healthy tissue, and this system is designed to systematically explore such effects in a laboratory setting. Multiple focused ultrasound transducers are connected to a programmable signal generator through an amplifier and a switch to select which transducer is active. Samples are placed beneath the transducer(s), using 3D-printed coupling cones filled with water to make a path for the ultrasound and set the depth of the focal spot within the target. This system is compatible with in vitro targets held in standard 6-well plates as well as with mouse flanks, through the assistance of a stereotactic frame. [[File:289982780-76b72cb4-2e9d-4a80-8a56-801b848d6971.png|frameless|600x600px]] Controller software allows for configuration of treatment parameters through a simple GUI application. With its default settings, the system can be configured to output focused ultrasound from the parameters shown in the table to the right. This allows for investigation of treatment parameters both within and well above the safety levels used for diagnostic ultrasound imaging. {| class="wikitable" |'''Parameter''' |'''Min''' |'''Max''' |'''Units''' |- |Frequency |70 |1000 |kHz |- |Output Voltage |0 |843 |V |- |Peak Negative Pressure |0 |1* |MPa |- |Mechanical Index |0 |3.8* | |- |ISPPA |0 |100* |W/cm2 |- |ISPTA |0 |1000* |mW/cm2 |- |Burst Length |2 |40 |ms |- |Duty Cycle |0.5 |10 |% |- |Treatment Duration |5 |60 * 15 |seconds |} ''*These are the maximum values that can be set when using the parameter as the target. If the parameter is derived and not used as the target, it may exceed this range to reach another parameter’s target value.'' Many of the components of the in vivo and in vitro systems were the same, although the precise fixturing used during the experiments was specific to each study. === Software Installation Instructions === Install/Run 32-bit Python 3.10+ (MUST BE 32-bit): `<path to 32 bit python>\python.exe install.py` This creates a virtual environment called env with the required dependencies. On Windows 10, Python may be installed to `C:\Users\<uname>\AppData\Local\Programs\Python\Python310-32` Install IVI Compliance Package 21.0 https://www.ni.com/en-us/support/downloads/drivers/download/packaged.ivi-compliance-package.409836.html Install DG4000 IVI Driver https://www.rigolna.com/products/waveform-generators/dg4000/ Install UltraSigma Instrument Connectivity Driver https://beyondmeasure.rigoltech.com/acton/attachment/1579/u-0003/0/-/-/-/-/ Enable Windows to load drivers from unknown sources: https://www.isunshare.com/windows-11/how-to-disable-driver-signature-enforcement-on-windows-11.html Download and install the Radiall USB Drivers from https://www.radiall.com/products/rf-microwave-switches/usb-coaxial-switches.html. Copy the folder containing `CP210xRuntime.dll`, `Radial_USBInterface.dll` and `Radial_USBInterface.xml` into a folder called `dll` in the root directory of the project. === Launch Software === Either double click on one of the <code>startapp.bat</code> files or run <code>env\Scripts\python.exe runapp.py <options></code> rtzifw5xvvkh1tyyamte7oszuis1r9g 177 176 2023-12-13T15:43:56Z OpenwaterPeterH 9 Added links to GitHub Repo 177 wikitext text/x-wiki == Preclinical Oncolysis System == === Overview === The Openwater Preclinical Oncolysis Prototype is designed to help researchers investigate the effect of different ultrasound parameters on a variety of in vitro and preclinical in vivo targets. Certain acoustic parameters may be well suited for damaging cancer cells while sparing surrounding healthy tissue, and this system is designed to systematically explore such effects in a laboratory setting. Multiple focused ultrasound transducers are connected to a programmable signal generator through an amplifier and a switch to select which transducer is active. Samples are placed beneath the transducer(s), using 3D-printed coupling cones filled with water to make a path for the ultrasound and set the depth of the focal spot within the target. This system is compatible with in vitro targets held in standard 6-well plates as well as with mouse flanks, through the assistance of a stereotactic frame. [[File:289982780-76b72cb4-2e9d-4a80-8a56-801b848d6971.png|frameless|600x600px]] Controller software allows for configuration of treatment parameters through a simple GUI application. With its default settings, the system can be configured to output focused ultrasound from the parameters shown in the table to the right. This allows for investigation of treatment parameters both within and well above the safety levels used for diagnostic ultrasound imaging. {| class="wikitable" |'''Parameter''' |'''Min''' |'''Max''' |'''Units''' |- |Frequency |70 |1000 |kHz |- |Output Voltage |0 |843 |V |- |Peak Negative Pressure |0 |1* |MPa |- |Mechanical Index |0 |3.8* | |- |ISPPA |0 |100* |W/cm2 |- |ISPTA |0 |1000* |mW/cm2 |- |Burst Length |2 |40 |ms |- |Duty Cycle |0.5 |10 |% |- |Treatment Duration |5 |60 * 15 |seconds |} ''*These are the maximum values that can be set when using the parameter as the target. If the parameter is derived and not used as the target, it may exceed this range to reach another parameter’s target value.'' Many of the components of the in vivo and in vitro systems were the same, although the precise fixturing used during the experiments was specific to each study. === Getting Started === Additional information can be found in the following repositories: * [https://github.com/OpenwaterHealth/opw_oncolysis_hw opw_oncolysis_hw] - assembly/operation instructions, mechanical drawings, 3D models * [https://github.com/OpenwaterHealth/opw_oncolysis_sw opw_oncolysis_sw] - software to configure and operate the system * [https://github.com/OpenwaterHealth/opw_oncolysis_data opw_oncolysis_data] - research data and experimental procedures === Software Installation Instructions === Install/Run 32-bit Python 3.10+ (MUST BE 32-bit): `<path to 32 bit python>\python.exe install.py` This creates a virtual environment called env with the required dependencies. On Windows 10, Python may be installed to `C:\Users\<uname>\AppData\Local\Programs\Python\Python310-32` Install IVI Compliance Package 21.0 https://www.ni.com/en-us/support/downloads/drivers/download/packaged.ivi-compliance-package.409836.html Install DG4000 IVI Driver https://www.rigolna.com/products/waveform-generators/dg4000/ Install UltraSigma Instrument Connectivity Driver https://beyondmeasure.rigoltech.com/acton/attachment/1579/u-0003/0/-/-/-/-/ Enable Windows to load drivers from unknown sources: https://www.isunshare.com/windows-11/how-to-disable-driver-signature-enforcement-on-windows-11.html Download and install the Radiall USB Drivers from https://www.radiall.com/products/rf-microwave-switches/usb-coaxial-switches.html. Copy the folder containing `CP210xRuntime.dll`, `Radial_USBInterface.dll` and `Radial_USBInterface.xml` into a folder called `dll` in the root directory of the project. === Launch Software === Either double click on one of the <code>startapp.bat</code> files or run <code>env\Scripts\python.exe runapp.py <options></code> jwyzaanfrnj1ftmld4ibx48q4q8127d 181 177 2023-12-13T16:35:37Z OpenwaterPeterH 9 Reformat Image 181 wikitext text/x-wiki == Preclinical Oncolysis System == === Overview === The Openwater Preclinical Oncolysis Prototype is designed to help researchers investigate the effect of different ultrasound parameters on a variety of in vitro and preclinical in vivo targets. Certain acoustic parameters may be well suited for damaging cancer cells while sparing surrounding healthy tissue, and this system is designed to systematically explore such effects in a laboratory setting. Multiple focused ultrasound transducers are connected to a programmable signal generator through an amplifier and a switch to select which transducer is active. Samples are placed beneath the transducer(s), using 3D-printed coupling cones filled with water to make a path for the ultrasound and set the depth of the focal spot within the target. This system is compatible with in vitro targets held in standard 6-well plates as well as with mouse flanks, through the assistance of a stereotactic frame. [[File:289982780-76b72cb4-2e9d-4a80-8a56-801b848d6971.png|right|frameless|600x600px|Block Diagram]] Controller software allows for configuration of treatment parameters through a simple GUI application. With its default settings, the system can be configured to output focused ultrasound from the parameters shown in the table to the right. This allows for investigation of treatment parameters both within and well above the safety levels used for diagnostic ultrasound imaging. {| class="wikitable" |'''Parameter''' |'''Min''' |'''Max''' |'''Units''' |- |Frequency |70 |1000 |kHz |- |Output Voltage |0 |843 |V |- |Peak Negative Pressure |0 |1* |MPa |- |Mechanical Index |0 |3.8* | |- |ISPPA |0 |100* |W/cm2 |- |ISPTA |0 |1000* |mW/cm2 |- |Burst Length |2 |40 |ms |- |Duty Cycle |0.5 |10 |% |- |Treatment Duration |5 |60 * 15 |seconds |} ''*These are the maximum values that can be set when using the parameter as the target. If the parameter is derived and not used as the target, it may exceed this range to reach another parameter’s target value.'' Many of the components of the in vivo and in vitro systems were the same, although the precise fixturing used during the experiments was specific to each study. === Getting Started === Additional information can be found in the following repositories: * [https://github.com/OpenwaterHealth/opw_oncolysis_hw opw_oncolysis_hw] - assembly/operation instructions, mechanical drawings, 3D models * [https://github.com/OpenwaterHealth/opw_oncolysis_sw opw_oncolysis_sw] - software to configure and operate the system * [https://github.com/OpenwaterHealth/opw_oncolysis_data opw_oncolysis_data] - research data and experimental procedures === Software Installation Instructions === Install/Run 32-bit Python 3.10+ (MUST BE 32-bit): `<path to 32 bit python>\python.exe install.py` This creates a virtual environment called env with the required dependencies. On Windows 10, Python may be installed to `C:\Users\<uname>\AppData\Local\Programs\Python\Python310-32` Install IVI Compliance Package 21.0 https://www.ni.com/en-us/support/downloads/drivers/download/packaged.ivi-compliance-package.409836.html Install DG4000 IVI Driver https://www.rigolna.com/products/waveform-generators/dg4000/ Install UltraSigma Instrument Connectivity Driver https://beyondmeasure.rigoltech.com/acton/attachment/1579/u-0003/0/-/-/-/-/ Enable Windows to load drivers from unknown sources: https://www.isunshare.com/windows-11/how-to-disable-driver-signature-enforcement-on-windows-11.html Download and install the Radiall USB Drivers from https://www.radiall.com/products/rf-microwave-switches/usb-coaxial-switches.html. Copy the folder containing `CP210xRuntime.dll`, `Radial_USBInterface.dll` and `Radial_USBInterface.xml` into a folder called `dll` in the root directory of the project. === Launch Software === Either double click on one of the <code>startapp.bat</code> files or run <code>env\Scripts\python.exe runapp.py <options></code> i7emsug3vebd0ggvtls6w42xwvwylrk 182 181 2023-12-13T16:35:53Z OpenwaterPeterH 9 182 wikitext text/x-wiki == Preclinical Oncolysis System == === Overview === The Openwater Preclinical Oncolysis Prototype is designed to help researchers investigate the effect of different ultrasound parameters on a variety of in vitro and preclinical in vivo targets. Certain acoustic parameters may be well suited for damaging cancer cells while sparing surrounding healthy tissue, and this system is designed to systematically explore such effects in a laboratory setting. Multiple focused ultrasound transducers are connected to a programmable signal generator through an amplifier and a switch to select which transducer is active. Samples are placed beneath the transducer(s), using 3D-printed coupling cones filled with water to make a path for the ultrasound and set the depth of the focal spot within the target. This system is compatible with in vitro targets held in standard 6-well plates as well as with mouse flanks, through the assistance of a stereotactic frame. [[File:289982780-76b72cb4-2e9d-4a80-8a56-801b848d6971.png|right|frameless|400x400px|Block Diagram]] Controller software allows for configuration of treatment parameters through a simple GUI application. With its default settings, the system can be configured to output focused ultrasound from the parameters shown in the table to the right. This allows for investigation of treatment parameters both within and well above the safety levels used for diagnostic ultrasound imaging. {| class="wikitable" |'''Parameter''' |'''Min''' |'''Max''' |'''Units''' |- |Frequency |70 |1000 |kHz |- |Output Voltage |0 |843 |V |- |Peak Negative Pressure |0 |1* |MPa |- |Mechanical Index |0 |3.8* | |- |ISPPA |0 |100* |W/cm2 |- |ISPTA |0 |1000* |mW/cm2 |- |Burst Length |2 |40 |ms |- |Duty Cycle |0.5 |10 |% |- |Treatment Duration |5 |60 * 15 |seconds |} ''*These are the maximum values that can be set when using the parameter as the target. If the parameter is derived and not used as the target, it may exceed this range to reach another parameter’s target value.'' Many of the components of the in vivo and in vitro systems were the same, although the precise fixturing used during the experiments was specific to each study. === Getting Started === Additional information can be found in the following repositories: * [https://github.com/OpenwaterHealth/opw_oncolysis_hw opw_oncolysis_hw] - assembly/operation instructions, mechanical drawings, 3D models * [https://github.com/OpenwaterHealth/opw_oncolysis_sw opw_oncolysis_sw] - software to configure and operate the system * [https://github.com/OpenwaterHealth/opw_oncolysis_data opw_oncolysis_data] - research data and experimental procedures === Software Installation Instructions === Install/Run 32-bit Python 3.10+ (MUST BE 32-bit): `<path to 32 bit python>\python.exe install.py` This creates a virtual environment called env with the required dependencies. On Windows 10, Python may be installed to `C:\Users\<uname>\AppData\Local\Programs\Python\Python310-32` Install IVI Compliance Package 21.0 https://www.ni.com/en-us/support/downloads/drivers/download/packaged.ivi-compliance-package.409836.html Install DG4000 IVI Driver https://www.rigolna.com/products/waveform-generators/dg4000/ Install UltraSigma Instrument Connectivity Driver https://beyondmeasure.rigoltech.com/acton/attachment/1579/u-0003/0/-/-/-/-/ Enable Windows to load drivers from unknown sources: https://www.isunshare.com/windows-11/how-to-disable-driver-signature-enforcement-on-windows-11.html Download and install the Radiall USB Drivers from https://www.radiall.com/products/rf-microwave-switches/usb-coaxial-switches.html. Copy the folder containing `CP210xRuntime.dll`, `Radial_USBInterface.dll` and `Radial_USBInterface.xml` into a folder called `dll` in the root directory of the project. === Launch Software === Either double click on one of the <code>startapp.bat</code> files or run <code>env\Scripts\python.exe runapp.py <options></code> ayvrkx3aj8ko0xljoomfrc5i3zimpys 183 182 2023-12-13T16:36:07Z OpenwaterPeterH 9 183 wikitext text/x-wiki == Preclinical Oncolysis System == === Overview === The Openwater Preclinical Oncolysis Prototype is designed to help researchers investigate the effect of different ultrasound parameters on a variety of in vitro and preclinical in vivo targets. Certain acoustic parameters may be well suited for damaging cancer cells while sparing surrounding healthy tissue, and this system is designed to systematically explore such effects in a laboratory setting.[[File:289982780-76b72cb4-2e9d-4a80-8a56-801b848d6971.png|right|frameless|400x400px|Block Diagram]] Multiple focused ultrasound transducers are connected to a programmable signal generator through an amplifier and a switch to select which transducer is active. Samples are placed beneath the transducer(s), using 3D-printed coupling cones filled with water to make a path for the ultrasound and set the depth of the focal spot within the target. This system is compatible with in vitro targets held in standard 6-well plates as well as with mouse flanks, through the assistance of a stereotactic frame. Controller software allows for configuration of treatment parameters through a simple GUI application. With its default settings, the system can be configured to output focused ultrasound from the parameters shown in the table to the right. This allows for investigation of treatment parameters both within and well above the safety levels used for diagnostic ultrasound imaging. {| class="wikitable" |'''Parameter''' |'''Min''' |'''Max''' |'''Units''' |- |Frequency |70 |1000 |kHz |- |Output Voltage |0 |843 |V |- |Peak Negative Pressure |0 |1* |MPa |- |Mechanical Index |0 |3.8* | |- |ISPPA |0 |100* |W/cm2 |- |ISPTA |0 |1000* |mW/cm2 |- |Burst Length |2 |40 |ms |- |Duty Cycle |0.5 |10 |% |- |Treatment Duration |5 |60 * 15 |seconds |} ''*These are the maximum values that can be set when using the parameter as the target. If the parameter is derived and not used as the target, it may exceed this range to reach another parameter’s target value.'' Many of the components of the in vivo and in vitro systems were the same, although the precise fixturing used during the experiments was specific to each study. === Getting Started === Additional information can be found in the following repositories: * [https://github.com/OpenwaterHealth/opw_oncolysis_hw opw_oncolysis_hw] - assembly/operation instructions, mechanical drawings, 3D models * [https://github.com/OpenwaterHealth/opw_oncolysis_sw opw_oncolysis_sw] - software to configure and operate the system * [https://github.com/OpenwaterHealth/opw_oncolysis_data opw_oncolysis_data] - research data and experimental procedures === Software Installation Instructions === Install/Run 32-bit Python 3.10+ (MUST BE 32-bit): `<path to 32 bit python>\python.exe install.py` This creates a virtual environment called env with the required dependencies. On Windows 10, Python may be installed to `C:\Users\<uname>\AppData\Local\Programs\Python\Python310-32` Install IVI Compliance Package 21.0 https://www.ni.com/en-us/support/downloads/drivers/download/packaged.ivi-compliance-package.409836.html Install DG4000 IVI Driver https://www.rigolna.com/products/waveform-generators/dg4000/ Install UltraSigma Instrument Connectivity Driver https://beyondmeasure.rigoltech.com/acton/attachment/1579/u-0003/0/-/-/-/-/ Enable Windows to load drivers from unknown sources: https://www.isunshare.com/windows-11/how-to-disable-driver-signature-enforcement-on-windows-11.html Download and install the Radiall USB Drivers from https://www.radiall.com/products/rf-microwave-switches/usb-coaxial-switches.html. Copy the folder containing `CP210xRuntime.dll`, `Radial_USBInterface.dll` and `Radial_USBInterface.xml` into a folder called `dll` in the root directory of the project. === Launch Software === Either double click on one of the <code>startapp.bat</code> files or run <code>env\Scripts\python.exe runapp.py <options></code> bxxgp5017v8oowss24fvpecilvdz1ey 297 183 2023-12-13T22:16:08Z OpenwaterPeterH 9 Remove info duplicated in repo 297 wikitext text/x-wiki == Preclinical Oncolysis System == === Overview === The Openwater Preclinical Oncolysis Prototype is designed to help researchers investigate the effect of different ultrasound parameters on a variety of in vitro and preclinical in vivo targets. Certain acoustic parameters may be well suited for damaging cancer cells while sparing surrounding healthy tissue, and this system is designed to systematically explore such effects in a laboratory setting.[[File:289982780-76b72cb4-2e9d-4a80-8a56-801b848d6971.png|right|frameless|400x400px|Block Diagram]] Multiple focused ultrasound transducers are connected to a programmable signal generator through an amplifier and a switch to select which transducer is active. Samples are placed beneath the transducer(s), using 3D-printed coupling cones filled with water to make a path for the ultrasound and set the depth of the focal spot within the target. This system is compatible with in vitro targets held in standard 6-well plates as well as with mouse flanks, through the assistance of a stereotactic frame. Controller software allows for configuration of treatment parameters through a simple GUI application. With its default settings, the system can be configured to output focused ultrasound from the parameters shown in the table to the right. This allows for investigation of treatment parameters both within and well above the safety levels used for diagnostic ultrasound imaging. {| class="wikitable" |'''Parameter''' |'''Min''' |'''Max''' |'''Units''' |- |Frequency |70 |1000 |kHz |- |Output Voltage |0 |843 |V |- |Peak Negative Pressure |0 |1* |MPa |- |Mechanical Index |0 |3.8* | |- |ISPPA |0 |100* |W/cm2 |- |ISPTA |0 |1000* |mW/cm2 |- |Burst Length |2 |40 |ms |- |Duty Cycle |0.5 |10 |% |- |Treatment Duration |5 |60 * 15 |seconds |} ''*These are the maximum values that can be set when using the parameter as the target. If the parameter is derived and not used as the target, it may exceed this range to reach another parameter’s target value.'' Many of the components of the in vivo and in vitro systems were the same, although the precise fixturing used during the experiments was specific to each study. === Getting Started === Additional information can be found in the following repositories: * [https://github.com/OpenwaterHealth/opw_oncolysis_hw opw_oncolysis_hw] - assembly/operation instructions, mechanical drawings, 3D models * [https://github.com/OpenwaterHealth/opw_oncolysis_sw opw_oncolysis_sw] - software to configure and operate the system * [https://github.com/OpenwaterHealth/opw_oncolysis_data opw_oncolysis_data] - research data and experimental procedures cu27df0k8lyk484rktofw02cst0g96p 395 297 2023-12-14T22:22:35Z Soren 11 395 wikitext text/x-wiki == Preclinical Oncolysis System == === Overview === The Openwater Preclinical Oncolysis Prototype is designed to help researchers investigate the effect of a wide range of ultrasound parameters on a variety of in vitro and preclinical in vivo targets. For example, certain acoustic parameters may be well suited for damaging cancer cells while sparing surrounding healthy tissue, and this system is designed to systematically explore such effects in a laboratory setting.[[File:289982780-76b72cb4-2e9d-4a80-8a56-801b848d6971.png|right|frameless|400x400px|Block Diagram]] Multiple focused ultrasound transducers are connected to a programmable signal generator through an amplifier and a switch to select which transducer is active. Samples are placed beneath the transducer(s), using 3D-printed coupling cones filled with water to make a path for the ultrasound and set the depth of the focal spot within the target. This system is compatible with in vitro targets held in standard 6-well plates as well as with mouse flanks, through the assistance of a stereotactic frame. Controller software allows for configuration of treatment parameters through a simple GUI application. With its default settings, the system can be configured to output focused ultrasound from the parameters shown in the table to the right. This allows for investigation of treatment parameters both within and well above the safety levels used for diagnostic ultrasound imaging. {| class="wikitable" |'''Parameter''' |'''Min''' |'''Max''' |'''Units''' |- |Frequency |70 |1000 |kHz |- |Output Voltage |0 |843 |V |- |Peak Negative Pressure |0 |1* |MPa |- |Mechanical Index |0 |3.8* | |- |ISPPA |0 |100* |W/cm2 |- |ISPTA |0 |1000* |mW/cm2 |- |Burst Length |2 |40 |ms |- |Duty Cycle |0.5 |10 |% |- |Treatment Duration |5 |60 * 15 |seconds |} ''*These are the maximum values that can be set when using the parameter as the target. If the parameter is derived and not used as the target, it may exceed this range to reach another parameter’s target value.'' Many of the components of the in vivo and in vitro systems were the same, although the precise fixturing used during the experiments was specific to each study. === Getting Started === Additional information can be found in the following repositories: * [https://github.com/OpenwaterHealth/opw_oncolysis_hw opw_oncolysis_hw] - assembly/operation instructions, mechanical drawings, 3D models * [https://github.com/OpenwaterHealth/opw_oncolysis_sw opw_oncolysis_sw] - software to configure and operate the system * [https://github.com/OpenwaterHealth/opw_oncolysis_data opw_oncolysis_data] - research data and experimental procedures eymoenxazrbksez0zo4of1v8t0pc6x2 632 395 2023-12-19T18:42:42Z Admin 1 Protected "[[Oncolysis]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 395 wikitext text/x-wiki == Preclinical Oncolysis System == === Overview === The Openwater Preclinical Oncolysis Prototype is designed to help researchers investigate the effect of a wide range of ultrasound parameters on a variety of in vitro and preclinical in vivo targets. For example, certain acoustic parameters may be well suited for damaging cancer cells while sparing surrounding healthy tissue, and this system is designed to systematically explore such effects in a laboratory setting.[[File:289982780-76b72cb4-2e9d-4a80-8a56-801b848d6971.png|right|frameless|400x400px|Block Diagram]] Multiple focused ultrasound transducers are connected to a programmable signal generator through an amplifier and a switch to select which transducer is active. Samples are placed beneath the transducer(s), using 3D-printed coupling cones filled with water to make a path for the ultrasound and set the depth of the focal spot within the target. This system is compatible with in vitro targets held in standard 6-well plates as well as with mouse flanks, through the assistance of a stereotactic frame. Controller software allows for configuration of treatment parameters through a simple GUI application. With its default settings, the system can be configured to output focused ultrasound from the parameters shown in the table to the right. This allows for investigation of treatment parameters both within and well above the safety levels used for diagnostic ultrasound imaging. {| class="wikitable" |'''Parameter''' |'''Min''' |'''Max''' |'''Units''' |- |Frequency |70 |1000 |kHz |- |Output Voltage |0 |843 |V |- |Peak Negative Pressure |0 |1* |MPa |- |Mechanical Index |0 |3.8* | |- |ISPPA |0 |100* |W/cm2 |- |ISPTA |0 |1000* |mW/cm2 |- |Burst Length |2 |40 |ms |- |Duty Cycle |0.5 |10 |% |- |Treatment Duration |5 |60 * 15 |seconds |} ''*These are the maximum values that can be set when using the parameter as the target. If the parameter is derived and not used as the target, it may exceed this range to reach another parameter’s target value.'' Many of the components of the in vivo and in vitro systems were the same, although the precise fixturing used during the experiments was specific to each study. === Getting Started === Additional information can be found in the following repositories: * [https://github.com/OpenwaterHealth/opw_oncolysis_hw opw_oncolysis_hw] - assembly/operation instructions, mechanical drawings, 3D models * [https://github.com/OpenwaterHealth/opw_oncolysis_sw opw_oncolysis_sw] - software to configure and operate the system * [https://github.com/OpenwaterHealth/opw_oncolysis_data opw_oncolysis_data] - research data and experimental procedures eymoenxazrbksez0zo4of1v8t0pc6x2 0 5 13 2023-12-12T21:28:16Z Gvigelet 4 Created page with "# Openwater Stroke Diagnosis Technology White Paper ## Background (defining stroke and problem statement) According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-c..." 13 wikitext text/x-wiki # Openwater Stroke Diagnosis Technology White Paper ## Background (defining stroke and problem statement) According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. ## Technology Overview ### Near-infrared window Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> <img src="wp_figure1.png" alt="Caption for Figure 1" width="432"> <br> <div style="width: 500px; margin: auto; text-align: center;"> <em>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</em> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> <img src="wp_figure2.png" alt="Caption for Figure 2" width="274"> <br> <div style="width: 400px; margin: auto; text-align: center;"> <em>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</em> </div> </p> ### Diffusion approximation The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> <img src="wp_figure3.png" alt="Caption for Figure 3" width="744"> <br> <div style="width: 750px; margin: auto; text-align: center;"> <em>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</em> </div> </p> ### Lasers, Speckle, and Blood Flow When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> <img src="wp_figure4.png" alt="Caption for Figure 4" width="471"> <br> <div style="width: 500px; margin: auto; text-align: center;"> <em>Figure 4: Photography of laser speckles from green laser light.</em> </div> </p> ## Prototype Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> <img src="wp_figure5.png" alt="Caption for Figure 5" width="686"> <br> <div style="width: 700px; margin: auto; text-align: center;"> <em>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</em> </div> </p> ## Initial Results ### Gas challenges Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> <img src="wp_figure6.png" alt="Caption for Figure 6" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <em>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</em> </div> </p> ### Temporary MCA occlusion We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> <img src="wp_figure7.png" alt="Caption for Figure 7" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <em>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</em> </div> </p> ### Permanent MCA occlusion We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> <img src="wp_figure8.png" alt="Caption for Figure 8" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <em>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</em> </div> </p> ### Permanent MCA occlusion monitored for 6 hours In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> <img src="wp_figure9.png" alt="Caption for Figure 9" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <em>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</em> </div> </p> ### Human forearm measurement We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> <img src="wp_figure10.png" alt="Caption for Figure 10" width="786"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <em>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</em> </div> </p> ## Summary We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. ## References 1. World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. 2. Saver, Jeffrey L., et al. "Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis." Jama 316.12 (2016): 1279-1289. 3. Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. 4. Drexler, Wolfgang, et al. "Optical coherence tomography today: speed, contrast, and multimodality." Journal of Biomedical Optics 19.7 (2014): 071412. 5. Vogel, Alfred, and Vasan Venugopalan. "Mechanisms of pulsed laser ablation of biological tissues." Chemical Reviews 103.2 (2003): 577-644. 6. Culver, J. P., et al. "Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging." Medical Physics 30.2 (2003): 235-247. 7. Yodh, A. G., and Chance, B. "Spectroscopy and imaging with diffusing light." Physics Today 48.3 (1995): 34-40. 8. Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. 9. Fercher, A. F., Briers, J. D. "Flow visualization by means of single-exposure speckle photography." Optics communications 37.5 (1981): 326-330. 10. Culver, J. P., et al. "Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest." Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. 11. Davis, Melissa F., Christopher Lay, and Ron D. Frostig. "Permanent cerebral vessel occlusion via double ligature and transection." JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 5bntfkqxkxsqqpznc8x1n0qszk4ai7i 14 13 2023-12-12T21:31:27Z Gvigelet 4 14 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology White Paper = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> <img src="wp_figure1.png" alt="Caption for Figure 1" width="432"> <br> <div style="width: 500px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])&lt;/em&gt;</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> <img src="wp_figure2.png" alt="Caption for Figure 2" width="274"> <br> <div style="width: 400px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])&lt;/em&gt;</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> <img src="wp_figure3.png" alt="Caption for Figure 3" width="744"> <br> <div style="width: 750px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.&lt;/em&gt;</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> <img src="wp_figure4.png" alt="Caption for Figure 4" width="471"> <br> <div style="width: 500px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 4: Photography of laser speckles from green laser light.&lt;/em&gt;</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> <img src="wp_figure5.png" alt="Caption for Figure 5" width="686"> <br> <div style="width: 700px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.&lt;/em&gt;</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> <img src="wp_figure6.png" alt="Caption for Figure 6" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.&lt;/em&gt;</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> <img src="wp_figure7.png" alt="Caption for Figure 7" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.&lt;/em&gt;</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> <img src="wp_figure8.png" alt="Caption for Figure 8" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.&lt;/em&gt;</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> <img src="wp_figure9.png" alt="Caption for Figure 9" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.&lt;/em&gt;</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> <img src="wp_figure10.png" alt="Caption for Figure 10" width="786"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.&lt;/em&gt;</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. jph6pkb7z6b7zh369syp3cam9bmcgne 17 14 2023-12-12T21:35:45Z Gvigelet 4 17 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology White Paper = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:wp_figure1.png Caption for Figure 1]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])&lt;/em&gt;</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> <img src="wp_figure2.png" alt="Caption for Figure 2" width="274"> <br> <div style="width: 400px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])&lt;/em&gt;</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> <img src="wp_figure3.png" alt="Caption for Figure 3" width="744"> <br> <div style="width: 750px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.&lt;/em&gt;</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> <img src="wp_figure4.png" alt="Caption for Figure 4" width="471"> <br> <div style="width: 500px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 4: Photography of laser speckles from green laser light.&lt;/em&gt;</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> <img src="wp_figure5.png" alt="Caption for Figure 5" width="686"> <br> <div style="width: 700px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.&lt;/em&gt;</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> <img src="wp_figure6.png" alt="Caption for Figure 6" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.&lt;/em&gt;</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> <img src="wp_figure7.png" alt="Caption for Figure 7" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.&lt;/em&gt;</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> <img src="wp_figure8.png" alt="Caption for Figure 8" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.&lt;/em&gt;</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> <img src="wp_figure9.png" alt="Caption for Figure 9" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.&lt;/em&gt;</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> <img src="wp_figure10.png" alt="Caption for Figure 10" width="786"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.&lt;/em&gt;</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. cv3l1wccucn7lfgxuzwl6qfklmfgfl7 21 17 2023-12-12T21:39:42Z Gvigelet 4 21 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology White Paper = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:wp_figure1.png Caption for Figure 1]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> <img src="wp_figure2.png" alt="Caption for Figure 2" width="274"> <br> <div style="width: 400px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])&lt;/em&gt;</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> <img src="wp_figure3.png" alt="Caption for Figure 3" width="744"> <br> <div style="width: 750px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.&lt;/em&gt;</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> <img src="wp_figure4.png" alt="Caption for Figure 4" width="471"> <br> <div style="width: 500px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 4: Photography of laser speckles from green laser light.&lt;/em&gt;</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> <img src="wp_figure5.png" alt="Caption for Figure 5" width="686"> <br> <div style="width: 700px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.&lt;/em&gt;</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> <img src="wp_figure6.png" alt="Caption for Figure 6" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.&lt;/em&gt;</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> <img src="wp_figure7.png" alt="Caption for Figure 7" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.&lt;/em&gt;</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> <img src="wp_figure8.png" alt="Caption for Figure 8" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.&lt;/em&gt;</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> <img src="wp_figure9.png" alt="Caption for Figure 9" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.&lt;/em&gt;</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> <img src="wp_figure10.png" alt="Caption for Figure 10" width="786"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>&lt;em&gt;Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.&lt;/em&gt;</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. hucfbi9zzjhogux47t4x8kv4fskzynu 22 21 2023-12-12T21:42:09Z Gvigelet 4 22 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology White Paper = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:wp_figure1.png Caption for Figure 1]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> <img src="wp_figure2.png" alt="Caption for Figure 2" width="274"> <br> <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> <img src="wp_figure3.png" alt="Caption for Figure 3" width="744"> <br> <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> <img src="wp_figure4.png" alt="Caption for Figure 4" width="471"> <br> <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> <img src="wp_figure5.png" alt="Caption for Figure 5" width="686"> <br> <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> <img src="wp_figure6.png" alt="Caption for Figure 6" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> <img src="wp_figure7.png" alt="Caption for Figure 7" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> <img src="wp_figure8.png" alt="Caption for Figure 8" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> <img src="wp_figure9.png" alt="Caption for Figure 9" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> <img src="wp_figure10.png" alt="Caption for Figure 10" width="786"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. t0rgor9x3n1phxoxlbe0szin79iq0f2 28 22 2023-12-12T21:57:49Z Gvigelet 4 28 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology White Paper = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png Caption for Figure 1]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> <img src="wp_figure2.png" alt="Caption for Figure 2" width="274"> <br> <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> <img src="wp_figure3.png" alt="Caption for Figure 3" width="744"> <br> <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> <img src="wp_figure4.png" alt="Caption for Figure 4" width="471"> <br> <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> <img src="wp_figure5.png" alt="Caption for Figure 5" width="686"> <br> <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> <img src="wp_figure6.png" alt="Caption for Figure 6" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> <img src="wp_figure7.png" alt="Caption for Figure 7" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> <img src="wp_figure8.png" alt="Caption for Figure 8" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> <img src="wp_figure9.png" alt="Caption for Figure 9" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> <img src="wp_figure10.png" alt="Caption for Figure 10" width="786"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 69a1rvhp64ovlu0tr2lfrhz04efyvr6 30 28 2023-12-12T21:58:07Z Gvigelet 4 30 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology White Paper = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> <img src="wp_figure2.png" alt="Caption for Figure 2" width="274"> <br> <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> <img src="wp_figure3.png" alt="Caption for Figure 3" width="744"> <br> <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> <img src="wp_figure4.png" alt="Caption for Figure 4" width="471"> <br> <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> <img src="wp_figure5.png" alt="Caption for Figure 5" width="686"> <br> <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> <img src="wp_figure6.png" alt="Caption for Figure 6" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> <img src="wp_figure7.png" alt="Caption for Figure 7" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> <img src="wp_figure8.png" alt="Caption for Figure 8" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> <img src="wp_figure9.png" alt="Caption for Figure 9" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> <img src="wp_figure10.png" alt="Caption for Figure 10" width="786"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. c8viu4g2etznxmzxsx311div99t5oxg 32 30 2023-12-12T21:58:53Z Gvigelet 4 32 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology White Paper = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> [[File:Wp_figure5.png]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> <img src="wp_figure6.png" alt="Caption for Figure 6" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> <img src="wp_figure7.png" alt="Caption for Figure 7" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> <img src="wp_figure8.png" alt="Caption for Figure 8" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> <img src="wp_figure9.png" alt="Caption for Figure 9" width="800"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> <img src="wp_figure10.png" alt="Caption for Figure 10" width="786"> <br> <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. bnl3u3xcv4ys4xajholeapvtlhq3ip8 41 32 2023-12-12T22:04:57Z Gvigelet 4 41 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology White Paper = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> [[File:Wp_figure5.png]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. rimf5bvq47helnlnbn5hplnhxm76ity 43 41 2023-12-12T22:07:11Z Gvigelet 4 43 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology White Paper = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> [[File:Wp_figure5.png]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 8xu8d70qrupz8ujgqr7pgf3i0j5okir 47 43 2023-12-12T22:09:06Z Gvigelet 4 47 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology White Paper = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> [[File:Wp_figure5.png]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. a89jt1orpd60wfx9t58huc6fmwygwaw 95 47 2023-12-13T02:16:24Z 24.92.36.30 /* Openwater Stroke Diagnosis Technology White Paper */ 95 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes [2]. At Openwater we are developing the fundamental technology to produce low-cost, portable medical imaging devices capable of both functional and structural imaging. Our first prototype measures tissue hemodynamics using near-infrared light. In particular, it is designed to measure differences in blood flow with the goal of reducing the time to diagnosis of ischemic stroke. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> [[File:Wp_figure5.png]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. ajjrh2aupzpozoa8aj62od7vc79vasg 96 95 2023-12-13T02:17:00Z 24.92.36.30 /* Background (defining stroke and problem statement) */ 96 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background (defining stroke and problem statement) == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> [[File:Wp_figure5.png]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. a40kvwh036xirss0p36anzinl0r0w8d 97 96 2023-12-13T02:17:18Z 24.92.36.30 /* Background (defining stroke and problem statement) */ 97 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our initial prototype, consists of a cart and a wand. A schematic is shown in Figure 5. The cart houses the laser, light detectors, computer, and various other optical and electronic components. A cord exiting from the side of the cart carries laser light to a wand. When the wand is pressed up against the patient and its acquisition button is pressed, a safe amount of near-infrared light is emitted and passes into the patient. Remitted light which carries information about the microvascular perfusion in the brain is collected by the wand’s detection fibers and transmitted to the detectors in the cart. For each subject, we plan to measure 12 locations (6 on each side of the head) with the wand. We expect each measurement to take approximately 7 seconds with an approximately equal amount of time required to move and reposition the wand between measurements. Thus a typical exam would take approximately 3 minutes. <p align="center"> [[File:Wp_figure5.png]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Schematic of the initial prototype. It consists of (A) a cart containing a laser, light detectors, computer, among other optical and electronic components, and (B) a wand which is connected to the cart by a cord. (C) Photograph of the wand. The next generation will be more compact, with the cart replaced by a small box which can be used in an ambulance.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 30s1ohapkstakc7py3ucqwg2hw08t9u 98 97 2023-12-13T02:19:23Z 24.92.36.30 /* Prototype */ 98 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Initial Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. ciuzxf3stnjqq4hdkt13xl45ugeum4m 100 98 2023-12-13T02:25:11Z Gvigelet 4 100 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 07ofjkk5l9igxhuls93p59trxr14ixp 101 100 2023-12-13T02:28:58Z Gvigelet 4 101 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. svaku5650wgl3yi5xehsjvvgydnh99a 102 101 2023-12-13T02:29:28Z Gvigelet 4 /* Large Vessel Occlusion (LVO) Detection */ 102 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. dm4kh9dwai47dtggmy0d0th93ct4ax4 103 102 2023-12-13T02:33:18Z Gvigelet 4 /* Large Vessel Occlusion (LVO) Detection */ 103 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. os3n1dbzot4d88ddgxhkwuog53f4i8w 105 103 2023-12-13T02:34:12Z Gvigelet 4 /* Large Vessel Occlusion (LVO) Detection */ 105 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 5rjsujqqwragpibxqsvk9bzf6coxxlp 106 105 2023-12-13T02:34:25Z Gvigelet 4 /* Summary */ 106 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> <span id="summary"></span> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 0vbqcarwkclo5umgmaba4axvvi2l577 107 106 2023-12-13T02:35:47Z Gvigelet 4 107 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. eg3oq0b9dd17jg4hrg77x6imixfrw6m 108 107 2023-12-13T02:36:23Z Gvigelet 4 108 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 0ntvh2kjmc76oxfwkzmdxfxopa30hti 109 108 2023-12-13T02:40:06Z Gvigelet 4 109 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow hardware, software, and AI repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 2gvawqib1eg0lh77w27eie660al274j 112 109 2023-12-13T02:44:27Z Gvigelet 4 112 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the blood flow hardware repository. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. t6vgvrvq95rj0762z1sus13u20l1yyo 113 112 2023-12-13T02:48:37Z Gvigelet 4 113 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. j0y12nm6wi0q9kraya828ahb3m414jf 114 113 2023-12-13T02:52:01Z Gvigelet 4 /* References */ 114 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.11,12 At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 rknsxrii1df24mfhmkalklgd1tdvlnd 115 114 2023-12-13T02:52:54Z Gvigelet 4 /* Openwater Stroke Diagnosis Technology */ 115 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 f7fhmhs2savgaus8b8hpl1nfmkio5uj 116 115 2023-12-13T02:53:58Z Gvigelet 4 /* References */ 116 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible (Goyal et al. 2016). Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays (Southerland et al., 2016). Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. c1mpo5izjk9324r82x6frrxfollkxzs 117 116 2023-12-13T02:56:11Z Gvigelet 4 /* Openwater Stroke Diagnosis Technology */ 117 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. nqse5mfy6zx361xi285doi88sukte78 121 117 2023-12-13T03:00:48Z Gvigelet 4 Gvigelet moved page [[Whitepaper]] to [[Openwater Stroke Diagnosis Technology]] 117 wikitext text/x-wiki <span id="openwater-stroke-diagnosis-technology-white-paper"></span> = Openwater Stroke Diagnosis Technology = <span id="background-defining-stroke-and-problem-statement"></span> == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. nqse5mfy6zx361xi285doi88sukte78 123 121 2023-12-13T03:01:12Z Gvigelet 4 /* Openwater Stroke Diagnosis Technology */ 123 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. lzxhg8s4v95li8vouivovukc6bhm06m 204 123 2023-12-13T17:20:15Z 24.92.36.30 /* Prototype */ 204 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> [[File:Wp_figure5b.mp4|913x913px]] </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. ixdvqtaffb539n5si58h64w7zneznr3 205 204 2023-12-13T17:22:49Z Gvigelet 4 /* Prototype */ 205 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. nzuxvo9plzj9usdgtj0lf9ffgjbu9k8 207 205 2023-12-13T17:25:14Z Gvigelet 4 /* Near-infrared window */ 207 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. 7cl4xsbyi1qa4cbotek4xuse1swc6a6 211 207 2023-12-13T17:31:27Z Gvigelet 4 /* Large Vessel Occlusion (LVO) Detection */ 211 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. 8hbz7cdnwljl66mftd3p7637j167zdz 212 211 2023-12-13T17:31:46Z Gvigelet 4 /* Validation with transcranial doppler ultrasound (TCD) */ 212 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. sdbuwkholfvp2u999lami48n1b23sgt 213 212 2023-12-13T17:31:55Z Gvigelet 4 /* Large Vessel Occlusion (LVO) Detection */ 213 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. <span id="technology-overview"></span> == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. rqtomy0xndx4uo44z4g6ugsfmicuh3v 216 213 2023-12-13T18:01:49Z OpenwaterEthan 7 216 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. gxwnjlxttrxh0xt8fsyhgeq3ixmhaxu 220 216 2023-12-13T19:09:28Z Gvigelet 4 /* Prototype */ 220 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|913x913px]][[File:speckleBFI.gif]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. sb44q3qcnouc3win7fjmlmbdqekx9bp 222 220 2023-12-13T19:11:11Z Gvigelet 4 222 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. lfqdiiyn87dh9787jy93dwcb949sfm9 224 222 2023-12-13T19:12:05Z Gvigelet 4 224 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. bjposk1p7mobo01gprvnz1jblmppfdi 225 224 2023-12-13T19:18:21Z Gvigelet 4 /* References */ 225 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. c2ozzumlapcehp66gvtitu8tkc26gxa 226 225 2023-12-13T19:22:24Z Gvigelet 4 /* Background */ 226 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled <ref>1</ref>. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. q6jnwz1njc7llo2fl3pqxcn1o1cwdr2 227 226 2023-12-13T19:22:45Z Gvigelet 4 /* References */ 227 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled <ref>1</ref>. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == <references> # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 0yylxikhbbnqb5vorem7d0suanjuegl 228 227 2023-12-13T19:23:16Z Gvigelet 4 /* References */ 228 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled <ref>1</ref>. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == <references> # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. </references> n8r9cqoyraclpb06q57ngofqd1eqo2a 229 228 2023-12-13T19:23:31Z Gvigelet 4 /* References */ 229 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled <ref>1</ref>. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. q6jnwz1njc7llo2fl3pqxcn1o1cwdr2 231 229 2023-12-13T19:25:21Z Gvigelet 4 231 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled <ref>1</ref>. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == <references /> # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 2rvm2382dgc5yrohbda53xhopbl2orp 232 231 2023-12-13T19:26:00Z Gvigelet 4 /* References */ 232 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled <ref>1</ref>. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[14]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[15]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[16]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[17]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[18]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[19]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[12],[13] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. q6jnwz1njc7llo2fl3pqxcn1o1cwdr2 236 232 2023-12-13T19:39:37Z Gvigelet 4 236 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [3]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [4]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [6]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. mx17cajgfth84nt2vfqaeiet8hbdg4a 238 236 2023-12-13T19:41:56Z Gvigelet 4 /* Near-infrared window */ 238 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [5])</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 3euj5o1kordbk70jik6c9ctrjval1qj 240 238 2023-12-13T19:49:26Z Gvigelet 4 /* Near-infrared window */ 240 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [7]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. lmvuilzprkf8dry9chvenitl4wuul36 241 240 2023-12-13T19:50:06Z Gvigelet 4 /* Diffusion approximation */ 241 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Saver, Jeffrey L., et al. “Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis.” Jama 316.12 (2016): 1279-1289. # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. b6xp8gopoy1k0ghvjtcow8ffycnik8f 242 241 2023-12-13T19:50:26Z Gvigelet 4 /* References */ 242 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [8]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [9]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. cel3sgs8ib64uoz40zzrlhzg207z36w 243 242 2023-12-13T19:51:19Z Gvigelet 4 /* Lasers, Speckle, and Blood Flow */ 243 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [10] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. oyfwl87tzf96hg2ixo9d6hi4hb0z192 244 243 2023-12-13T19:52:30Z Gvigelet 4 /* Gas challenges */ 244 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [11]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. pv7kkwv27mpgzuro15369kt9tahm7ok 245 244 2023-12-13T19:52:48Z Gvigelet 4 /* Permanent MCA occlusion */ 245 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:Wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 1gzesfm4hoa3ejwxqgb7jpn6y4kr7m0 246 245 2023-12-13T19:55:40Z Gvigelet 4 /* Lasers, Speckle, and Blood Flow */ 246 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. ej1xcdeoaj6hgjk6nrj0gzf3ofx9222 247 246 2023-12-13T19:56:52Z Gvigelet 4 /* Lasers, Speckle, and Blood Flow */ 247 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg Link to Laser Speckle Green Image] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 69e1sxi6kyskf4e334bjui1o6izlqs9 248 247 2023-12-13T19:57:27Z Gvigelet 4 /* Lasers, Speckle, and Blood Flow */ 248 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> <gallery> https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg|Description of the image </gallery> <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. f5e5jwcga7zidk2ljndvijb3qludd9o 249 248 2023-12-13T19:58:51Z Gvigelet 4 /* Lasers, Speckle, and Blood Flow */ 249 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg|Description of the image </gallery> <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light.</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. k2trjgp06n2etksvojw03ujtz6fg24y 250 249 2023-12-13T19:59:19Z Gvigelet 4 /* Lasers, Speckle, and Blood Flow */ 250 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. bwy3ju2pmy8sfzpwdcjbdx4qeds97i4 274 250 2023-12-13T21:39:43Z 24.92.36.30 274 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 LVO detection] publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our publication. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. iixkl7y1n7qeeh6x6omcvrs63qpoi4f 277 274 2023-12-13T21:41:18Z 24.92.36.30 /* Validation with transcranial doppler ultrasound (TCD) */ 277 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 LVO detection] publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 publication]. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 3fdynlglsghx9i565imnkcu97pmveym 279 277 2023-12-13T21:41:36Z 24.92.36.30 /* Background */ 279 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 publication]. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == We have developed a new technique for measuring tissue hemodynamics centimeters below the surface of living tissue. The technique uses short laser pulses of near-infrared light with very narrow bandwidth to measure microvascular perfusion, as well as the optical absorption of hemoglobin. We have conducted small animal studies which demonstrate our ability to distinguish between rats that have, and have not, undergone ligation of their middle cerebral artery. We are currently constructing a portable cart-based device capable of measuring stroke patients in a clinical setting. Initial testing indicates that the device is capable of measuring small blood flow changes at depths required to interrogate the human brain. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 9ki4jen2nw4qdq5foaeerg9elwwuz96 283 279 2023-12-13T21:44:12Z 24.92.36.30 /* Summary */ 283 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 publication]. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Large vessel occlusion study results.</pre> </div> </p> == Summary == Detecting the likely presence of  LVO stroke before patients are transported to hospitals is a huge unmet need.  Solving this critical problem has the potential to prevent death and disabilities for many thousands of people every year.  Openwater has built a new instrument for measuring tissue hemodynamics centimeters below the surface of living tissue with the potential to solve this problem. The instrument uses short laser pulses of near-infrared light with very narrow bandwidth to measure blood flow. We have conducted preclinical and clinical studies demonstrating the sensitivity of the device to small changes in blood flow.  In an initial hospital study, the device was able to distinguish between subjects with and without LVO who had been brought to the hospital with suspected strokes. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. tew08hv6h7b6sny0z8o00wmxgtrgueo 312 283 2023-12-13T23:04:30Z Gvigelet 4 /* Large Vessel Occlusion (LVO) Detection */ 312 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 publication]. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13a.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13a: Sensitivity and specificity are reported for each diagnostic tool. The reported PPV and NPV are based on an estimated LVO prevalence of 5% and 10% in a prehospital setting, and the reported FPs and NPs are based on a sample of 1000 patients. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. PPV indicates positive predictive value. NPV indicates negative predictive value. FP indicates number of false positive. FN indicates number of false negative.</pre> </div> </p> == Summary == Detecting the likely presence of  LVO stroke before patients are transported to hospitals is a huge unmet need.  Solving this critical problem has the potential to prevent death and disabilities for many thousands of people every year.  Openwater has built a new instrument for measuring tissue hemodynamics centimeters below the surface of living tissue with the potential to solve this problem. The instrument uses short laser pulses of near-infrared light with very narrow bandwidth to measure blood flow. We have conducted preclinical and clinical studies demonstrating the sensitivity of the device to small changes in blood flow.  In an initial hospital study, the device was able to distinguish between subjects with and without LVO who had been brought to the hospital with suspected strokes. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. iqdijpuyxov6n35w98vxeumo3ychxp1 314 312 2023-12-13T23:07:04Z Gvigelet 4 /* Large Vessel Occlusion (LVO) Detection */ 314 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 publication]. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13a.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13a: Sensitivity and specificity are reported for each diagnostic tool. The reported PPV and NPV are based on an estimated LVO prevalence of 5% and 10% in a prehospital setting, and the reported FPs and NPs are based on a sample of 1000 patients. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. PPV indicates positive predictive value. NPV indicates negative predictive value. FP indicates number of false positive. FN indicates number of false negative.</pre> </div> </p> == Summary == Detecting the likely presence of  LVO stroke before patients are transported to hospitals is a huge unmet need.  Solving this critical problem has the potential to prevent death and disabilities for many thousands of people every year.  Openwater has built a new instrument for measuring tissue hemodynamics centimeters below the surface of living tissue with the potential to solve this problem. The instrument uses short laser pulses of near-infrared light with very narrow bandwidth to measure blood flow. We have conducted preclinical and clinical studies demonstrating the sensitivity of the device to small changes in blood flow.  In an initial hospital study, the device was able to distinguish between subjects with and without LVO who had been brought to the hospital with suspected strokes. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. q0lyrqhjm3pj4973nis2na91ecsso5x 315 314 2023-12-13T23:08:00Z Gvigelet 4 /* Large Vessel Occlusion (LVO) Detection */ 315 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 publication]. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13a.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Sensitivity and specificity are reported for each diagnostic tool. The reported PPV and NPV are based on an estimated LVO prevalence of 5% and 10% in a prehospital setting, and the reported FPs and NPs are based on a sample of 1000 patients. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. PPV indicates positive predictive value. NPV indicates negative predictive value. FP indicates number of false positive. FN indicates number of false negative.</pre> </div> </p> [[File:Wp_figure13bc.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13ab: ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale.</pre> </div> </p> == Summary == Detecting the likely presence of  LVO stroke before patients are transported to hospitals is a huge unmet need.  Solving this critical problem has the potential to prevent death and disabilities for many thousands of people every year.  Openwater has built a new instrument for measuring tissue hemodynamics centimeters below the surface of living tissue with the potential to solve this problem. The instrument uses short laser pulses of near-infrared light with very narrow bandwidth to measure blood flow. We have conducted preclinical and clinical studies demonstrating the sensitivity of the device to small changes in blood flow.  In an initial hospital study, the device was able to distinguish between subjects with and without LVO who had been brought to the hospital with suspected strokes. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. nfhtmz0pgyiyj1zw8zxzdgit5h622fd 316 315 2023-12-13T23:08:12Z Gvigelet 4 /* Large Vessel Occlusion (LVO) Detection */ 316 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 publication]. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see our publication. <p align="center"> [[File:Wp_figure13a.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Sensitivity and specificity are reported for each diagnostic tool. The reported PPV and NPV are based on an estimated LVO prevalence of 5% and 10% in a prehospital setting, and the reported FPs and NPs are based on a sample of 1000 patients. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. PPV indicates positive predictive value. NPV indicates negative predictive value. FP indicates number of false positive. FN indicates number of false negative.</pre> </div> </p> <p align="center"> [[File:Wp_figure13bc.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13ab: ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale.</pre> </div> </p> == Summary == Detecting the likely presence of  LVO stroke before patients are transported to hospitals is a huge unmet need.  Solving this critical problem has the potential to prevent death and disabilities for many thousands of people every year.  Openwater has built a new instrument for measuring tissue hemodynamics centimeters below the surface of living tissue with the potential to solve this problem. The instrument uses short laser pulses of near-infrared light with very narrow bandwidth to measure blood flow. We have conducted preclinical and clinical studies demonstrating the sensitivity of the device to small changes in blood flow.  In an initial hospital study, the device was able to distinguish between subjects with and without LVO who had been brought to the hospital with suspected strokes. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. 73vcnuwkaf6l2pc1mzobxwa5xgxk8d6 509 316 2023-12-15T21:42:05Z KedarGrama 6 /* Large Vessel Occlusion (LVO) Detection */ 509 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 publication]. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 our publication]. <p align="center"> [[File:Wp_figure13a.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Sensitivity and specificity are reported for each diagnostic tool. The reported PPV and NPV are based on an estimated LVO prevalence of 5% and 10% in a prehospital setting, and the reported FPs and NPs are based on a sample of 1000 patients. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. PPV indicates positive predictive value. NPV indicates negative predictive value. FP indicates number of false positive. FN indicates number of false negative.</pre> </div> </p> <p align="center"> [[File:Wp_figure13bc.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13ab: ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale.</pre> </div> </p> == Summary == Detecting the likely presence of  LVO stroke before patients are transported to hospitals is a huge unmet need.  Solving this critical problem has the potential to prevent death and disabilities for many thousands of people every year.  Openwater has built a new instrument for measuring tissue hemodynamics centimeters below the surface of living tissue with the potential to solve this problem. The instrument uses short laser pulses of near-infrared light with very narrow bandwidth to measure blood flow. We have conducted preclinical and clinical studies demonstrating the sensitivity of the device to small changes in blood flow.  In an initial hospital study, the device was able to distinguish between subjects with and without LVO who had been brought to the hospital with suspected strokes. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. kpi63k4nuv4h879xjzup3ynuj5jetwa 633 509 2023-12-19T18:42:57Z Admin 1 Protected "[[Openwater Stroke Diagnosis Technology]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 509 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 publication]. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 our publication]. <p align="center"> [[File:Wp_figure13a.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Sensitivity and specificity are reported for each diagnostic tool. The reported PPV and NPV are based on an estimated LVO prevalence of 5% and 10% in a prehospital setting, and the reported FPs and NPs are based on a sample of 1000 patients. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. PPV indicates positive predictive value. NPV indicates negative predictive value. FP indicates number of false positive. FN indicates number of false negative.</pre> </div> </p> <p align="center"> [[File:Wp_figure13bc.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13ab: ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale.</pre> </div> </p> == Summary == Detecting the likely presence of  LVO stroke before patients are transported to hospitals is a huge unmet need.  Solving this critical problem has the potential to prevent death and disabilities for many thousands of people every year.  Openwater has built a new instrument for measuring tissue hemodynamics centimeters below the surface of living tissue with the potential to solve this problem. The instrument uses short laser pulses of near-infrared light with very narrow bandwidth to measure blood flow. We have conducted preclinical and clinical studies demonstrating the sensitivity of the device to small changes in blood flow.  In an initial hospital study, the device was able to distinguish between subjects with and without LVO who had been brought to the hospital with suspected strokes. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. kpi63k4nuv4h879xjzup3ynuj5jetwa 670 633 2024-01-03T00:51:32Z KedarGrama 6 670 wikitext text/x-wiki == Background == According to the World Health Organization, every year 15 million people suffer a stroke, 5 million of whom die, and another 5 million of which are disabled [1]. In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)[2]. Due to the rapid loss of brain tissue during stroke, prompt diagnosis and treatment is critical for improving stroke outcomes[3]. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial circulation account for up to 46% of AIS and are considered refractory to intravenous blood thinner tissue plasminogen activator (tPA)[4]. Emergency transportation for endovascular therapy has now become the standard of care for anterior LVO resulting, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements when comparing endovascular treatment to medical therapy alone[5]. However, EVT is only offered at a limited number of advanced endovascular capable centers (ECCs), which makes access to this highly beneficial therapy difficult for patients who are outside the catchment area of an ECC. Given the finding that >55% of patients in EVT RCTs had poor long-term outcomes of severe morbidity or death, whereas >90% of patients had good neurological outcomes with minimal to no deficits if they received EVT within 2.5 hours of onset, there is a critical need to get patients to an ECC as soon as possible[6]. Currently however patients are routed to closest hospital or primary stroke center for initial workup, which often is not capable of providing EVT, resulting in significant delay in care due to interfacility transfer time delays[7]. Early LVO recognition during pre-hospital care presents an opportunity to route patients to endovascular capable centers and thereby reduce treatment times and improve outcomes. In fact, the American Heart Association along with its Mission: Lifeline® Stroke algorithm recommend emergency health services (EMS) route high-likelihood LVO patients to a comprehensive or thrombectomy capable stroke center depending on the additional transportation time.[8],[9] At Openwater we have developed a low-cost, portable prototype that measures tissue hemodynamics using near-infrared light with unprecedented sensitivity. The pages below give a general overview of the technology, a description of the prototype that we have deployed, and show some of our initial results. More detail about the device can be found in the blood flow [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw hardware], [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_sw software], and [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_ai AI] repositories. Clinical results can be found in our clinical validation and LVO detection publications. == Technology Overview == <span id="near-infrared-window"></span> === Near-infrared window === Biomedical optics is a rapidly expanding field which is providing biologists and clinicians new ways to detect, diagnose, and study disease. However, most optical techniques can only be used to gain information near the tissue surface [10]. For example, confocal microscopy is only capable of imaging up to 50 μm below the tissue surface, and even optical coherence tomography only images at depth of up to 2-3 mm in opaque tissues. The two principal obstacles to looking deep into tissue with visible light are the high degree of light absorption and scattering by tissue. There exists however, a spectral region in the near infrared (NIR) where the absorption of light by tissue is relatively low. As can be seen in Figure 1, the absorption of light by oxy- and deoxy-hemoglobin drops dramatically at around 600 nm. Likewise, the absorption of light by water is very low through wavelengths up to around 900 nm. As a result, there is a window in the NIR from about 650-950 nm where light can penetrate more deeply into tissue, and much work has been done using NIR light with the techniques mentioned above in order to maximize their penetration depth. However, the second obstacle, the high degree of light scattering, limits the use of any of the above techniques to a few millimeters [11]. The mean free path (the average distance a photon will travel before interacting with matter) of visible light in tissue is only about 100 μm, and multiple scattering events will cause the direction of the average photon to be randomized after about 1 mm. Thus, techniques which rely on ballistic or quasi-ballistic light (aka time-of-flight) are inherently limited in depth. <p align="center"> [[File:Wp_figure1.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 1: Near-infrared absorption spectrum of light absorbing chromophores in tissue demonstrating the spectral window from 650 - 950 nm where light can penetrate to depths of many centimeters. (Adapted from [12])</pre> </div> </p> The problem is not that NIR light cannot penetrate deeply into tissue. It does. For example, NIR light transmitted through 10 cm of human breast tissue can be detected [13]. The problem is that beyond a few millimeters deep, almost all of the remaining photons have been scattered multiple times, and their directions are random. For example, Figure 2 is a cell phone photograph of the author with an LED in his mouth. The light exiting the tissue is easily detected even though the LED itself is not visible due to the light scattering. In order to acquire information about tissues deep below the surface, a method is needed which permits this information to be extracted from detected photons which have been scattered many times. Techniques using scattered near-infrared light to interrogate deep tissues are often referred to as Diffuse Optics and/or Near-Infrared Spectroscopy (NIRS). <p align="center"> [[File:Wp_figure2.png]] <div style="width: 400px; margin: auto; text-align: center;"> <pre>Figure 2: Cell phone photograph of the author with an LED in his mouth. Red light is easily detected even though the LED itself is not visible.</pre> </div> </p> <span id="diffusion-approximation"></span> === Diffusion approximation === The key observation underlying diffuse optical methods is that the paths of NIR photons in tissue can be described as a random walk with a step length equal to the distance over which their directions become randomized [14]. As mentioned above, this distance is about 1 mm in human tissue. As a result, if experiments are performed over distances much greater than the step length, the propagation of the NIR photons can be modeled as a diffusive process. Figure 3 is a schematic describing this process. It depicts light from an optical fiber being injected into tissue. The light intensity decreases with distance from the position of the light source. A set of detector fibers collects light exiting the tissue at various points. Using an accurate model of how the light propagates, it is possible to use the detected light to gain information about the tissue through which it has passed, including the concentration of absorbing chromophores such as oxy- and deoxy-hemoglobin, the amount of light scattering, and the concentration and lifetime of exogenous fluorophores. <p align="center"> [[File:Wp_figure3.png]] <div style="width: 750px; margin: auto; text-align: center;"> <pre>Figure 3: Schematic of the diffusion of light through tissue. The photons scatter randomly and spread out causing the light intensity to decrease with distance from the source. This process is well described using mathematical models of diffusion. Some of the remitted light is collected by optical fibers and is used to determine the optical properties of the tissue.</pre> </div> </p> <span id="lasers-speckle-and-blood-flow"></span> === Lasers, Speckle, and Blood Flow === When laser light is reflected from a rough surface and then detected (e.g. by your retina or a camera) the resulting image contains randomly located light and dark spots commonly referred to as speckle [15]. The light and dark spots are due to the constructive and destructive interference between light waves that travel different distances. This phenomenon can be readily observed by shining a laser pointer at a wall and observing the reflected light (see Figure 4). It also occurs when laser light passes through highly scattering media, such as biological tissue. If the light scattering particles which compose the scattering media are in motion, the locations of constructive and destructive interference of the light waves (i.e. the speckle) change in time. If the change from bright to dark occurs on a time scale equal to or shorter to the exposure time of the light detector, the contrast of the speckle (i.e. the difference between bright and dark spots) decreases. As a result, the contrast of the speckle pattern is related to the motion of the interfaces scattering the light. More motion, due to either the scatterers moving faster, or more of the scatterers moving, results in a decrease in speckle contrast [16]. Openwater’s blood flow technology combines diffuse optics with measurements of laser speckles. Short pulses of monochromatic laser light are injected into tissue using a fiber optic. The light diffuses through the tissue. Some of the light remitted from the tissue is collected by fiber optics located at various locations. The contrast of the measured speckles is related to the number of moving red blood cells and their speed. A measure of microvascular blood perfusion is then calculated using an algorithm based on a mathematical model of the relationship between scatterer motion and laser speckles. <p align="center"> [[File:wp_figure4.png]] <div style="width: 500px; margin: auto; text-align: center;"> <pre>Figure 4: Photography of laser speckles from green laser light. (https://commons.wikimedia.org/wiki/File:Laser_speckle_green.jpg)</pre> </div> </p> <span id="prototype"></span> == Prototype == Our portable blood flow device uses our custom laser combined with CMOS cameras to measure deep within the body. For more information about our blood flow measurement hardware, see the [https://github.com/OpenwaterHealth/opw_bloodflow_gen2_hw blood flow hardware repository]. <p align="center"> [[File:Wp_figure5.png|594x594px]][[File:speckleBFI.gif|360x360px]] <div style="width: 700px; margin: auto; text-align: center;"> <pre>Figure 5: Photograph of the blood flow measurement prototype, schematic of the light path through the body, and sample data.</pre> </div> </p> <span id="initial-results"></span> == Preclinical Results == <span id="gas-challenges"></span> === Gas challenges === Our initial small animal in vivo measurements were designed to detect global changes in hemodynamics caused by the inhalation of varying gas mixtures. Anesthetized rats (0.5 L/min air flow with 2% isoflurane) were given hypercapnic and hypoxic gas challenges. The hypercapnic challenge consisted of increasing the CO 2 in the gas mixture from 0% to 5% for 30 seconds. During the hypoxic challenge, the O 2 in the inhaled mixture was reduced from 20% to 10% for a period of 30 seconds. Typical results are shown in Figure 6. The left column of graphs shows the results for optical attenuation (Figure 6B) and blood flow (Figure 6C) of a hypercapnic challenge. Dilation of blood vessels due to the inhalation of excess levels of CO 2 results in increases in both optical attenuation (less light is detected) and blood flow. The right column shows the results of a hypoxic challenge. As the rat is deprived of oxygen, a decrease in hemoglobin concentration leads to a decrease in optical attenuation (more light is detected) [17] as shown in Figure 6D. No overall change in blood flow is observed (Figure 6E). In all the graphs the blue lines represent the raw time curves which are then smoothed to produce the orange curves. The oscillations in the raw blood flow curves are due to the pulse of the rat, which is easily measured. <p align="center"> [[File:Wp figure6.png|868x868px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 6: Results from the gas challenges of a rat. (A) Picture of a rat showing a small wand pressed up against its head. Light travels to and from the wand via the optical fibers shown. (B-C) Results of the hypercapnic challenge which causes the dilation of blood vessels. (B) Light attenuation increases (due to increased blood volume) during the challenge. (C) Blood flow also increases. (D-E) Results of the hypoxic challenge. (D) Decreased oxygenation of hemoglobin results in a decrease in optical attenuation. (E) No change in the rate of blood flow is observed during the hypoxic challenge.</pre> </div> </p> <span id="temporary-mca-occlusion"></span> === Temporary MCA occlusion === We monitored rats during temporary occlusions of their middle cerebral artery (MCA). For each rat, a small hole was bored in the rear portion of the right side of the skull exposing MCA. Once MCA was exposed, we began monitoring blood flow by injecting light into the front right of the head, and detecting the remitted light with three detectors located at the right rear, middle, and left front of the head. After two minutes of data collection, we used a microvascular clip to occlude MCA. The clip remained in place for 1 minute before being removed, and the rat was monitored for an additional 2 minutes. Results from a representative rat are shown in Figure 7. Application of the clip resulted in an immediate increase in speckle contrast, which returns to normal as soon as the clip is removed. Furthermore, the increase in speckle contrast was greatest for the detector on the right side of the head which probed the right hemisphere of the brain. In contrast, the detector on the front left of the head, which probed the front of the brain, measured a much smaller difference. <p align="center"> [[File:Wp_figure7.png|832x832px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 7: Results of a temporary middle cerebral artery occlusion. A hole is bored into the rear right side of the skull exposing the right middle cerebral artery (MCA). Light is injected into the rat at the right front portion of the head and is detected at 3 locations. After 2 minutes of data acquisition, a micro-surgical clip is used to occlude MCA for 1 minute. Decreased blood flow leads to an increase in speckle contrast. The increase is largest on the detector located on the right side, and smallest for the detector on the left.</pre> </div> </p> <span id="permanent-mca-occlusion"></span> === Permanent MCA occlusion === We performed a study to determine if we could reliably detect left vs. right hemisphere blood flow differences after occluding the MCA. Rats received a permanent occlusion of their right MCA following previously established methods [18]. Measurements were taken on the left and right sides of the head both before and after surgery. A total of six rats were measured, out of which 10 baseline pairs of left/right measurements and 5 post stroke pairs of left/right measurements were made. (One rat died during surgery, and two rats received multiple baseline measurements). On average, left to right differences were 7x larger after occlusion (see Figure 8). In addition, the smallest post occlusion difference (0.06) in any of the rats was twice as large as the largest baseline difference (0.03) among all the rats, demonstrating that the measurement could be used to determine if a rat received the occlusion. <p align="center"> [[File:Wp_figure8.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 8: Results of the permanent middle cerebral artery (MCA) occlusion study. The left and right sides of 6 rats were measured both before and after permanent MCA occlusion. Differences in speckle contrast between the left and right hemispheres were on average 7x greater after occlusion than before.</pre> </div> </p> <span id="permanent-mca-occlusion-monitored-for-6-hours"></span> === Permanent MCA occlusion monitored for 6 hours === In order to ensure the measured changes in blood flow post occlusion were not transient, two additional rats were monitored for 6 hours after surgery. In both cases the measured difference between hemispheres persisted for the duration of the experiment, albeit with a 20% reduction as shown in Figure 9. As is typical in our experiments, the surgical procedure itself, in which MCA is both exposed and perturbed, leads to an increase in speckle contrast (decrease in blood flow) on the exposed side, while the speckle contrast on the contralateral side remains constant. (Only the difference between left and right hemispheres is shown in Figure 9.) Also visible are changes in the speckle contrast that occur when the concentration of isoflurane (and thus blood flow) is changed. <p align="center"> [[File:Wp_figure9.png|800x800px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 9: Results of an extended post occlusion measurement. Differences in speckle contrast persist up to 6 hours after occlusion. Data was acquired using three different laser pulse widths corresponding to the three curves.</pre> </div> </p> <span id="clinical-results"></span> == Clinical Results== <span id="human-forearm-measurement"></span> === Human forearm measurement === We used our device to monitor the blood flow in the forearm of a healthy human subject. The forearm was monitored for 90 seconds. After a 30 second baseline, a blood pressure cuff was inflated on the upper portion of the arm. Pressure was maintained for approximately 30 seconds, then the cuff was deflated and the monitoring was continued for another 30 seconds. As shown in Figure 10, the pulse is clearly visible during the initial and final 30 seconds, when the cuff is not inflated. As the cuff is inflated, there is a transient increase in measured blood flow which may be due to the subject’s motion. For the duration of the occlusion, the pulse is no longer visible, and the mean blood flow value is lower. <p align="center"> [[File:Wp_figure10.png]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 10: Measured blood flow in the forearm of a healthy human subject. The subject’s pulse is easily seen in the blue line. The black line is a low pass filtered version of the blue line. At approximately 30 seconds a blood pressure cuff is inflated for 30 seconds, reducing blood flow to the forearm. During the occlusion, the pulse disappears and the mean blood flow value is reduced.</pre> </div> </p> <span id="validation-with-transcranial-doppler-ultrasound"></span> === Validation with transcranial doppler ultrasound (TCD) === In order to validate the Openwater blood flow measurement device, we made simultaneous measurements with transcranial doppler ultrasound . 25 subjects were asked to hold there breath, and the resulting changes in cerebral blood flow were measured. Major results are shown here. For more information see our [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 publication]. <p align="center"> [[File:Wp_figure11.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 11: Changes in blood flow measured with the Openwater device were highly correlated with transcranial doppler ultrasound.</pre> </div> </p> <p align="center"> [[File:Wp_figure12.png|929x929px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 12: The measured waveforms from Openwater & transcranial doppler ultrasound had similar morphology.</pre> </div> </p> <span id="large-vessel-occlusion-detection"></span> === Large Vessel Occlusion (LVO) Detection === The Openwater headset outperformed prehospital stroke scales for LVO detection in patients who underwent acute stroke evaluation at two comprehensive stroke centers. The device demonstrated a higher sensitivity, specificity, and area under the curve. Further investigation is needed to assess the performance in prehospital settings. For more information, see the [[Blood Flow Gen 2 LVO Classification and Analysis|machine learning wiki]] and [https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 our publication]. <p align="center"> [[File:Wp_figure13a.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13: Sensitivity and specificity are reported for each diagnostic tool. The reported PPV and NPV are based on an estimated LVO prevalence of 5% and 10% in a prehospital setting, and the reported FPs and NPs are based on a sample of 1000 patients. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale. PPV indicates positive predictive value. NPV indicates negative predictive value. FP indicates number of false positive. FN indicates number of false negative.</pre> </div> </p> <p align="center"> [[File:Wp_figure13bc.png|808x808px]] <div style="width: 800px; margin: auto; text-align: center;"> <pre>Figure 13ab: ROC analysis for LVO detection: (A) The receiver operator characteristic area under the curve is depicted when using raw scores. The area under the curve for the Openwater optical blood flow monitor is larger than that of LAMS. (B) The receiver operator characteristic area under the curve is depicted when using thresholded scores. The Openwater threshold was ≥ 0.80. The RACE threshold was ≥ 5. The LAMS threshold was ≥ 4. ROC indicates receiver operator characteristic. LVO indicates large vessel occlusion. RACE indicates the rapid arterial occlusion evaluation scale. LAMS indicates the Los Angeles Motor Scale.</pre> </div> </p> == Summary == Detecting the likely presence of  LVO stroke before patients are transported to hospitals is a huge unmet need.  Solving this critical problem has the potential to prevent death and disabilities for many thousands of people every year.  Openwater has built a new instrument for measuring tissue hemodynamics centimeters below the surface of living tissue with the potential to solve this problem. The instrument uses short laser pulses of near-infrared light with very narrow bandwidth to measure blood flow. We have conducted preclinical and clinical studies demonstrating the sensitivity of the device to small changes in blood flow.  In an initial hospital study, the device was able to distinguish between subjects with and without LVO who had been brought to the hospital with suspected strokes. <span id="references"></span> == References == # World Health Organization. The World Health Report 2002: Reducing Risks, Promoting Healthy Life. World Health Organization, 2002. # Virani SS et al; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association. Circulation. 2020 Mar 3;141(9):e139-e596 # Silva GS, Nogueira RG. Endovascular Treatment of Acute Ischemic Stroke. Continuum (Minneap Minn). 2020 Apr;26(2):310-331. # Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. Neurosurgery, 85(suppl_1), S4–S8. # McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In Scientific World Journal (Vol. 2019). # Goyal M,et al., Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016 Apr 23;387(10029):1723-31. # Southerland AM, Johnston KC, Molina CA, Selim MH, Kamal N, Goyal M. Suspected Large Vessel Occlusion: Should Emergency Medical Services Transport to the Nearest Primary Stroke Center or Bypass to a Comprehensive Stroke Center With Endovascular Capabilities? Stroke. 2016 Jul;47(7):1965-7. # Stroke AHAML. Emergency Medical Services Acute Stroke Routing. https://www.heart.org/en/professional/quality-improvement/mission-lifeline/mission-lifeline-stroke. 2020. Accessed November 1, 2023. # Jauch EC,, et al., Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society. Stroke. 2021;52:e133-e152 # Vo-Dinh, Tuan, 2nd ed. Biomedical Photonics Handbook. CRC Press, 2014. # Drexler, Wolfgang, et al. “Optical coherence tomography today: speed, contrast, and multimodality.” Journal of Biomedical Optics 19.7 (2014): 071412. # Vogel, Alfred, and Vasan Venugopalan. “Mechanisms of pulsed laser ablation of biological tissues.” Chemical Reviews 103.2 (2003): 577-644. # Culver, J. P., et al. “Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: Evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging.” Medical Physics 30.2 (2003): 235-247. # Yodh, A. G., and Chance, B. “Spectroscopy and imaging with diffusing light.” Physics Today 48.3 (1995): 34-40. # Goodman, Joseph W. Speckle Phenomena in Optics: Theory and Applications . Roberts and Company Publishers, 2007. # Fercher, A. F., Briers, J. D. “Flow visualization by means of single-exposure speckle photography.” Optics communications 37.5 (1981): 326-330. # Culver, J. P., et al. “Diffuse optical measurement of hemoglobin and cerebral blood flow in rat brain during hypercapnia, hypoxia and cardiac arrest.” Oxygen Transport to Tissue XXIII. Springer, Boston, MA, 2003. 293-297. # Davis, Melissa F., Christopher Lay, and Ron D. Frostig. “Permanent cerebral vessel occlusion via double ligature and transection.” JoVE (Journal of Visualized Experiments) 77 (2013): e50418. llurlnfwv52zwisizozn3ixdly8beqv 0 123 642 2023-12-20T00:35:24Z Openwaterpete 5 Created page with "In order to reduce cost and size in a generic transducer driver and costom compact and low cost drived has been created and tested. This removes a very expensve genreal device and replaces it with a battery powered specific design" 642 wikitext text/x-wiki In order to reduce cost and size in a generic transducer driver and costom compact and low cost drived has been created and tested. This removes a very expensve genreal device and replaces it with a battery powered specific design 9w2jjoed1919ifizdxujcowjynrzu0y 644 642 2023-12-20T00:37:56Z Openwaterpete 5 addeded first comments 644 wikitext text/x-wiki [[File:Head and ustx.png|thumb|Figure 1 headset and minautrized USTX driver]] In order to reduce cost and size in a generic transducer driver and costom compact and low cost drived has been created and tested. This removes a very expensve genreal device and replaces it with a battery powered specific design tn342nzn9xfzg9wplnv3ajpzp8fgbsc 647 644 2023-12-20T00:44:31Z Openwaterpete 5 fig 2 647 wikitext text/x-wiki [[File:Head and ustx.png|thumb|Figure 1 headset and minautrized USTX driver]] [[File:Ustx box photo.png|thumb|figure 2]] In order to reduce cost and size in a generic transducer driver and costom compact and low cost drived has been created and tested. This removes a very expensve genreal device and replaces it with a battery powered specific design rzi1jprkzo9xxiah5es5vz5wqbnd13d 649 647 2023-12-20T00:45:48Z Openwaterpete 5 649 wikitext text/x-wiki [[File:Head and ustx.png|thumb|Figure 1 headset and minautrized USTX driver]] [[File:Ustx box photo.png|thumb|figure 2]] [[File:Ustx model.png|thumb|Figure 3]] In order to reduce cost and size in a generic transducer driver and costom compact and low cost drived has been created and tested. This removes a very expensve genreal device and replaces it with a battery powered specific design 3p1a2zs7rsi9x4ksr5xxclrlt57o36o 652 649 2023-12-20T16:34:11Z Openwaterpete 5 652 wikitext text/x-wiki [[File:Head and ustx.png|thumb|Figure 1 headset and minautrized USTX driver]] [[File:Ustx box photo.png|thumb|figure 2]] [[File:Ustx model.png|thumb|Figure 3]] In order to reduce cost and size in a generic transducer driver and costom compact and low cost drived has been created and tested. This removes a very expensve general transducer driving device and replaces it with a battery powered specific design t40mevjkggnbw770m8cuuvo1ndeeq1h 653 652 2023-12-20T16:45:44Z Openwaterpete 5 653 wikitext text/x-wiki [[File:Head and ustx.png|thumb|Figure 1 headset and minautrized USTX driver]] [[File:Ustx box photo.png|thumb|figure 2]] [[File:Ustx model.png|thumb|Figure 3]] The initial prototype design includes a very expensive and large general transducer driving system, this system is impractical for a market system. The Gen2 design removed the most expensive component and replaced it with a custom battery driven transducer driver to prove feasibility of such a system. A low cost driver has been created and tested in the Gen 2 system. The learning with this device design informs the Gen 3 general market design which is in processs. bc2rlqmx3qkyfq303cuo9qr5l4y75he 654 653 2023-12-20T22:36:41Z KedarGrama 6 KedarGrama moved page [[USTX Board]] to [[Openwater Ultrasound Transmit Module (USTX)]]: Descriptive title that Chris Bawiec demanded 653 wikitext text/x-wiki [[File:Head and ustx.png|thumb|Figure 1 headset and minautrized USTX driver]] [[File:Ustx box photo.png|thumb|figure 2]] [[File:Ustx model.png|thumb|Figure 3]] The initial prototype design includes a very expensive and large general transducer driving system, this system is impractical for a market system. The Gen2 design removed the most expensive component and replaced it with a custom battery driven transducer driver to prove feasibility of such a system. A low cost driver has been created and tested in the Gen 2 system. The learning with this device design informs the Gen 3 general market design which is in processs. bc2rlqmx3qkyfq303cuo9qr5l4y75he 658 654 2023-12-21T00:00:18Z Chrisbawiec 14 rewrote the wiki page 658 wikitext text/x-wiki [[File:Head and ustx.png|thumb|Figure 1 headset and minautrized USTX driver]] [[File:Ustx box photo.png|thumb|figure 2]] [[File:Ustx model.png|thumb|Figure 3]] Overview The USTX module is a work-in-progress device developed by Openwater to provide low-cost, standalone ultrasound transmission capabilities for the Open-LIFU platform. It offers beamforming configuration and control for driving 128-channel ultrasound transducers. Key Features * Beamforming Control: ** Accepts commands from a host PC to configure beamforming parameters. ** Manages analog front-end beamformer devices for precise control of transducer firing sequences. ** Enables focusing of 128-channel transducers to specific focal points using customized patterns or waveforms. * Power Options: ** Powered by a battery for semi-portable operation, though with limited output voltage. ** Can be connected to a programmable DC power supply for higher output voltage (up to ±100V). ** Supports PWM (pulse-width modulation) for per-element power control and apodization, leading to better focusing and reduced unwanted sound waves (grating and side lobes). * Pulse Parameter Adjustment: ** Offers flexibility in adjusting pulse parameters beyond frequency and pressure amplitude. ** Allows customization of pulse length, pulse repetition frequency, burst duration, burst length, and total time. Physical Characteristics * Housed in a 3D-printed case for portability. Applications While still under development, potential applications for the USTX module include: * Low-intensity Focused Ultrasound (LIFU) therapy * Research and development in various ultrasound-related fields Additional Information * For more technical details, please refer to the USTX documentation provided by Openwater. * The module's development status and future availability can be inquired about through Openwater's official channels. b121tesrk8u7r10ar7f13epxpufi9ba 659 658 2023-12-21T00:02:41Z KedarGrama 6 659 wikitext text/x-wiki [[File:Head and ustx.png|thumb|Figure 1 headset and minautrized USTX driver]] [[File:Ustx box photo.png|thumb|figure 2]] [[File:Ustx model.png|thumb|Figure 3]] == Overview == The USTX module is a work-in-progress device developed by Openwater to provide low-cost, standalone ultrasound transmission capabilities for the Open-LIFU platform. It offers beamforming configuration and control for driving 128-channel ultrasound transducers. == Key Features == * Beamforming Control: ** Accepts commands from a host PC to configure beamforming parameters. ** Manages analog front-end beamformer devices for precise control of transducer firing sequences. ** Enables focusing of 128-channel transducers to specific focal points using customized patterns or waveforms. * Power Options: ** Powered by a battery for semi-portable operation, though with limited output voltage. ** Can be connected to a programmable DC power supply for higher output voltage (up to ±100V). ** Supports PWM (pulse-width modulation) for per-element power control and apodization, leading to better focusing and reduced unwanted sound waves (grating and side lobes). * Pulse Parameter Adjustment: ** Offers flexibility in adjusting pulse parameters beyond frequency and pressure amplitude. ** Allows customization of pulse length, pulse repetition frequency, burst duration, burst length, and total time. == Physical Characteristics == * Housed in a 3D-printed case for portability. == Applications == While still under development, potential applications for the USTX module include: * Low-intensity Focused Ultrasound (LIFU) therapy * Research and development in various ultrasound-related fields == Additional Information == * For more technical details, please refer to the USTX documentation provided by Openwater. * The module's development status and future availability can be inquired about through Openwater's official channels. 9rlqyvm2bf8z72xim4i0xtwb8zxlm49 660 659 2023-12-21T01:28:36Z Chrisbawiec 14 added link to repo 660 wikitext text/x-wiki [[File:Head and ustx.png|thumb|Figure 1 headset and minautrized USTX driver]] [[File:Ustx box photo.png|thumb|figure 2]] [[File:Ustx model.png|thumb|Figure 3]] == Overview == The USTX module is a work-in-progress device developed by Openwater to provide low-cost, standalone ultrasound transmission capabilities for the Open-LIFU platform. It offers beamforming configuration and control for driving 128-channel ultrasound transducers. == Key Features == * Beamforming Control: ** Accepts commands from a host PC to configure beamforming parameters. ** Manages analog front-end beamformer devices for precise control of transducer firing sequences. ** Enables focusing of 128-channel transducers to specific focal points using customized patterns or waveforms. * Power Options: ** Powered by a battery for semi-portable operation, though with limited output voltage. ** Can be connected to a programmable DC power supply for higher output voltage (up to ±100V). ** Supports PWM (pulse-width modulation) for per-element power control and apodization, leading to better focusing and reduced unwanted sound waves (grating and side lobes). * Pulse Parameter Adjustment: ** Offers flexibility in adjusting pulse parameters beyond frequency and pressure amplitude. ** Allows customization of pulse length, pulse repetition frequency, burst duration, burst length, and total time. == Physical Characteristics == * Housed in a 3D-printed case for portability. == Applications == While still under development, potential applications for the USTX module include: * Low-intensity Focused Ultrasound (LIFU) therapy * Research and development in various ultrasound-related fields == Additional Information == * For more technical details, please refer to the [https://github.com/OpenwaterHealth/opw_ustx USTX documentation] provided by Openwater. * The module's development status and future availability can be inquired about through Openwater's official channels. nems0chlv6bk6p80n8l79oc8dlb7ymm 0 1 1 2023-12-12T19:10:16Z MediaWiki default 2 1 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 22vz5zlxa2zctewimaum2bf1due8hkl 9 1 2023-12-12T21:22:39Z Gvigelet 4 /* Getting started */ 9 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. [Technology/WhitePaper:Technology WhitePaper] == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 4umvnxyz6hpwmlutkxdkg4i9fhrubjo 10 9 2023-12-12T21:24:37Z Gvigelet 4 10 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. [http://http://162.246.254.83/index.php?title=WhitePaper:Technology WhitePaper] == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 7fjs9wvnqredewu5pcf39f9g02s98vp 11 10 2023-12-12T21:26:24Z Gvigelet 4 11 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. [http://http://162.246.254.83/index.php?title=WhitePaper WhitePaper] == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] dwscfql0yhan9l3tajhiei5vjh0ujc7 15 11 2023-12-12T21:32:57Z Gvigelet 4 15 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. [http://162.246.254.83/index.php/WhitePaper WhitePaper] == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 0up968qc7utrlo4ii7x1zaz2ec5hd4g 16 15 2023-12-12T21:33:27Z Gvigelet 4 16 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. [http://162.246.254.83/index.php/Whitepaper WhitePaper] == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] epgkfhjwa80lx1wipncvycaiaczvt3x 18 16 2023-12-12T21:35:51Z Openwaterpete 5 18 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. [http://162.246.254.83/index.php/Whitepaper WhitePaper] [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 4bcpf3s4t12ju5wx29m3k1tumf78isx 19 18 2023-12-12T21:36:46Z Openwaterpete 5 19 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. [http://162.246.254.83/index.php/Whitepaper WhitePaper] * blood folw gen 2 hardware [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] pj8al08ll9140dxs2o6rcjdnqyn3qdl 20 19 2023-12-12T21:37:38Z Openwaterpete 5 20 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. [http://162.246.254.83/index.php/Whitepaper WhitePaper] * blood folw gen 2 hardware [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 6c3bz4lqm4up3uzqr8o8zgnmyz4neai 23 20 2023-12-12T21:43:25Z Gvigelet 4 23 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * blood folw gen 2 hardware [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] p9rq0xe3n8p7ypj9r2x56nuevdhesbf 24 23 2023-12-12T21:43:39Z Gvigelet 4 24 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * blood folw gen 2 hardware [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 9jijn7ap8md8l8o4gu12dzfdljgftad 26 24 2023-12-12T21:57:06Z 50.227.118.138 26 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] == [['''Blood flow gen 2 hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware]]]== == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] t2bb49yt8h4c3w6dqvki7bcemtwrev1 29 26 2023-12-12T21:57:50Z 50.227.118.138 29 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] == '''Blood flow gen 2 hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware]== == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] drlmb3qgm3slqqui7lk51app88jkko8 31 29 2023-12-12T21:58:31Z 50.227.118.138 /* Blood flow gen 2 hardware: [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] */ 31 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] == '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware]== == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 0ha941u1ureqi6h4hwjwyhnthbx835r 36 31 2023-12-12T21:59:56Z 50.227.118.138 36 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] == '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware]== == '''Blood Flow Gen 1 Hardware''': == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] g6gv4d1baaz92lyebqrptlbsx5khgxg 38 36 2023-12-12T22:01:02Z 50.227.118.138 38 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] == '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware]== == '''Blood Flow Gen 1 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_1_Hardware]== == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] esfopytpz9xbzv8qig1z1ss5pgr72tl 40 38 2023-12-12T22:02:35Z 50.227.118.138 40 wikitext text/x-wiki <strong>MediaWiki has been installed.</strong> Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] == '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware]== == '''Blood Flow Gen 1 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_1_Hardware]== == '''Phantoms''': [http://162.246.254.83/index.php/Phantoms]== == Getting started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 9f526poomhzg2uq4guzl73zj6o7jjco 48 40 2023-12-12T22:25:29Z 50.227.118.138 48 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == External Links == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] * '''Blood Flow Gen 1 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_1_Hardware] * '''Phantoms''': [http://162.246.254.83/index.php/Phantoms] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] iyc7ymscs2la4i08htaxx2u7rhzobuk 53 48 2023-12-12T22:56:18Z Admin 1 added a link to the camera test rig in the registry 53 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == External Links == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] * '''Blood Flow Gen 1 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_1_Hardware] * '''Phantoms''': [http://162.246.254.83/index.php/Phantoms] * [[Camera Test Rig]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 2ec6uazdp6j61rq5hxfjcs0l18lmxt8 54 53 2023-12-12T22:56:51Z Admin 1 54 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] * '''Blood Flow Gen 1 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_1_Hardware] * '''Phantoms''': [http://162.246.254.83/index.php/Phantoms] * [[Camera Test Rig]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] sz61st5y47d72hgw0n5r7wd6v1r6ak1 59 54 2023-12-12T23:22:16Z Opw12 8 /* Table of Contents */ 59 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] * '''Blood Flow Gen 1 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_1_Hardware] * '''Phantoms''': [http://162.246.254.83/index.php/Phantoms] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 1u4vyshtqnett12rgxaogazzbygm581 71 59 2023-12-12T23:49:17Z Openwaterpete 5 71 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] * '''Blood Flow Gen 1 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_1_Hardware] * '''Phantoms''': [http://162.246.254.83/index.php/Phantoms] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] h4a9i7ifq9bgrvb45prpmjdtept0g0m 76 71 2023-12-12T23:58:36Z Openwaterpete 5 76 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] * [[Blood Flow Gen 1 Hardware]] '''Blood Flow Gen 1 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_1_Hardware] * '''Phantoms''': [http://162.246.254.83/index.php/Phantoms] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 5we5xzw5pvdarl2izbuf6pab1t6godc 77 76 2023-12-12T23:59:12Z Openwaterpete 5 77 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] * [[Blood Flow Gen 1 Hardware]] * '''Phantoms''': [http://162.246.254.83/index.php/Phantoms] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 08ehxky5ttz2r7c71x9o5ytsh1d0qyg 78 77 2023-12-12T23:59:40Z Openwaterpete 5 78 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * '''Blood Flow Gen 2 Hardware''': [http://162.246.254.83/index.php/Blood_Flow_Gen_2_Hardware] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] qpd7uo9k3n5yu3to40kbvquvz5zvho2 79 78 2023-12-13T00:00:10Z Openwaterpete 5 79 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] rfd1xrn9f7xkp3d4ont3n1cbdxkppe9 94 79 2023-12-13T02:13:55Z 24.92.36.30 94 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology White Paper] * [[Theory of Operation]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] guae3kabi4twn4ug48mpi8pm9pi0gh7 118 94 2023-12-13T02:58:15Z Gvigelet 4 /* Table of Contents */ 118 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [http://162.246.254.83/index.php/Whitepaper Openwater Stroke Diagnosis Technology] * [[Theory of Operation]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] o630xdtgbypzxovnemiv20hbo0c64l9 119 118 2023-12-13T02:59:07Z Gvigelet 4 /* Table of Contents */ 119 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [[Whitepaper] Openwater Stroke Diagnosis Technology] * [[Theory of Operation]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 9vel7us0uaiueg2yho5iy5rwstse9rj 120 119 2023-12-13T02:59:44Z Gvigelet 4 /* Table of Contents */ 120 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [[Whitepaper]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] s02gefbgof9eyayd75uwys6gdl7zauk 124 120 2023-12-13T03:01:33Z Gvigelet 4 /* Table of Contents */ 124 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] lj3pvqsa2plvw2m3zdi77q14satca1v 125 124 2023-12-13T03:02:42Z Gvigelet 4 /* Table of Contents */ 125 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 1qdnr7wvmfnwvgpa6ddcvyurffpg5ga 133 125 2023-12-13T03:22:37Z Gvigelet 4 133 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] fvpcta6gqi3owy9srd1bx7mhq90ubf7 188 133 2023-12-13T16:48:09Z OpenwaterPeterH 9 Added links to Oncolysis and Neuromodulation 188 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Table of Contents == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] * [[Oncolysis]] * [[Neuromodulation]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] pbmc96medugb0mjeyr4dj6uh8ua02j5 230 188 2023-12-13T19:24:12Z 69.181.108.2 230 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Developing technologies with the potential to revolutionize patient care.= We are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. == Table of Contents == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] * [[Oncolysis]] * [[Neuromodulation]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] mgwvs7g8915fothyvg768m1xo4r26nc 233 230 2023-12-13T19:26:32Z 69.181.108.2 /* Developing technologies with the potential to revolutionize patient care. */ 233 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. == Table of Contents == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] * [[Oncolysis]] * [[Neuromodulation]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] qa1bli7w4t6il7gptmlvet3t3d7u1xg 234 233 2023-12-13T19:30:16Z 69.181.108.2 234 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entier Openwater technical database to help create an open source community freely distribuing non-invasive diagonostica and therapeutic technologies. == Table of Contents == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] * [[Oncolysis]] * [[Neuromodulation]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] fuxn40m3sf0jj5rwsyfmxikbmr5w3gm 235 234 2023-12-13T19:30:58Z 69.181.108.2 235 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Table of Contents == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] * [[Phantoms]] * [[Camera Test Rig]] * [[Acousto Optic|Acousto-Optic System]] * [[Laser Characterization]] * [[Oncolysis]] * [[Neuromodulation]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 6tskteb7bd9qoft1uemn55vbufbj1m3 237 235 2023-12-13T19:40:57Z Opw12 8 Converted list of projects to different headings 237 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == == Regulatory == == Peer Reviewed Publications == == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] r1905ynyd169ra92iyvf09f1wylb054 239 237 2023-12-13T19:43:08Z Opw12 8 /* Patents */ 239 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == Patent page to include direct link to our Patent Pledge document on GitHub. == Regulatory == == Peer Reviewed Publications == == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 0re1dbghyblauag0uft0odtxh5m8nk3 255 239 2023-12-13T20:33:40Z Opw12 8 /* Peer Reviewed Publications */ 255 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == Patent page to include direct link to our Patent Pledge document on GitHub. == Regulatory == == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] nowqqd0bt9v44kwim861fgt40ik6psp 258 255 2023-12-13T20:39:33Z Opw12 8 /* Regulatory */ 258 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == Patent page to include direct link to our Patent Pledge document on GitHub. == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] rvaef6pjpzh8q02v8j4l0lf6fz8kbyi 261 258 2023-12-13T20:42:41Z Opw12 8 /* Patents */ 261 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] c35s51brmqu2t0fleul5ot58jnzu600 263 261 2023-12-13T21:21:03Z 69.181.108.2 263 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwter Health- Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] e0dss56mqpusk1xsw15yt71t902ow0x 264 263 2023-12-13T21:22:24Z 69.181.108.2 264 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =**Openwter Health** - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] p4fdw9ohqnsgwwi39mmeqs357i35ctz 265 264 2023-12-13T21:23:38Z 69.181.108.2 265 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =** Openwter Health ** - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] fhw37o3hiab7vpzs46aemcruml1ed63 266 265 2023-12-13T21:23:59Z 69.181.108.2 266 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwter Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] jtflmy7u4zqrc5nm87vsd3elecpkvwg 267 266 2023-12-13T21:25:28Z 69.181.108.2 267 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =OpenwterHealth - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] gv5kp95ezzh1l5hbqrzhnk4830wt68d 282 267 2023-12-13T21:43:48Z 69.181.108.2 282 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =OpenwaterHealth - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wikipage is to summarize the entire Openwater technical database to help create an open source community freely distribuing our non-invasive diagonostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] lynof53mmkvrbwinbr768eydh3sybi4 295 282 2023-12-13T21:58:37Z KedarGrama 6 295 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database to help create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] mju4v2nxkvcwuh931l51yxwn959mwnl 298 295 2023-12-13T22:27:11Z 69.181.108.2 298 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] pycr0pa78m1k1tf3oluurgr9oe0fnmh 299 298 2023-12-13T22:27:54Z 69.181.108.2 299 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == [[Neuromodulation]] == Oncolysis == [[Oncolysis]] == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] e49yppg936jyf5dyjxxp3x8pfs1jhb8 300 299 2023-12-13T22:29:39Z Opw12 8 300 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == [[Patents]] == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 1rzr9rra6h5k2quowm2icdmbejyqnpt 303 300 2023-12-13T22:34:59Z Opw12 8 /* Patents */ 303 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == [[Regulatory]] == Peer Reviewed Publications == Peer reviewed [[publications]] == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] n72zflo2z4fk8hcy3dx9e4rurd8pb0f 306 303 2023-12-13T22:41:04Z Opw12 8 /* Regulatory */ 306 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == [[Regulatory]] == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 29ct4pqmfogjctcoyfddlk8657d7qzj 307 306 2023-12-13T22:42:50Z Opw12 8 /* Regulatory */ 307 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == * [[Phantoms]] * [[Camera Test Rig]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] olhkderpcq7qzfzbmvmsolzg8juhaf8 308 307 2023-12-13T22:48:31Z Opw12 8 /* Laboratory Testing */ 308 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Acousto-Optic Imaging == [[Acousto Optic|Acousto-Optic System]] == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] aht19tb1bps6hesm60uuxhxn7qk25se 309 308 2023-12-13T22:53:19Z Opw12 8 /* Acousto-Optic Imaging */ 309 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Acousto-Optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == Getting Started == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] gmwmjvojuav0e0bcnlom34skxgfzrli 310 309 2023-12-13T23:02:50Z OpenwaterEthan 7 310 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == * [[Blood Flow Gen 1 Software]] * [[Blood Flow Gen 1 Hardware]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Acousto-Optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == MediaWiki Usage Guide == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] momufq89kilkya057xammnct02ylh73 311 310 2023-12-13T23:02:54Z Opw12 8 /* Blood Flow Gen 1 */ 311 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Hardware]] == Blood Flow Gen 1 == The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Acousto-Optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == MediaWiki Usage Guide == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 81ej0abtnvxakrha7o1e4yyl9kh5r31 320 311 2023-12-13T23:11:02Z Opw12 8 /* Blood Flow Gen 2 */ 320 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. =Openwater Health - Developing technologies with the potential to revolutionize patient care.= At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] == Blood Flow Gen 1 == The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Acousto-Optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == MediaWiki Usage Guide == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] aroy23u8f9aluy20syev3ikqnkkdf8p 322 320 2023-12-13T23:14:55Z Opw12 8 322 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Openwater Health - Developing technologies with the potential to revolutionize patient care. == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow Gen 2 == The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Whitepaper|Openwater Stroke Diagnosis Technology]] * [[Theory of Operation]] * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] == Blood Flow Gen 1 == The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Acousto-Optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == MediaWiki Usage Guide == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 11h86v10zzc2e4wrx2ehm0paxh5h9l8 326 322 2023-12-13T23:54:54Z 50.227.118.138 /* Openwater Health - Developing technologies with the potential to revolutionize patient care. */ 326 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Openwater Health - Developing technologies with the potential to revolutionize patient care. == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow == need blurb here * [[Whitepaper|Openwater Stroke Diagnosis Technology]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Acousto-Optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == MediaWiki Usage Guide == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] r4yizf6umorqtxo70jqfpmq3nwrpz8f 327 326 2023-12-13T23:58:01Z 50.227.118.138 327 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Openwater Health - Developing technologies with the potential to revolutionize patient care. == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow == need blurb here * [[Whitepaper|Openwater Stroke Diagnosis Technology Review]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Acousto-Optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == MediaWiki Usage Guide == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] ecv63smpiiybkikzo839s9k8rt6ddm0 333 327 2023-12-14T00:14:46Z 50.227.118.138 333 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Openwater Health - Developing technologies with the potential to revolutionize patient care. == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow == need blurb here * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Acousto-Optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == MediaWiki Usage Guide == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] 7ds1oue9xnfiou30u8p3jd55meyvqyo 335 333 2023-12-14T00:23:58Z Opw12 8 /* Acousto-Optic Imaging */ 335 wikitext text/x-wiki Consult the [https://www.mediawiki.org/wiki/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software. == Openwater Health - Developing technologies with the potential to revolutionize patient care. == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow == need blurb here * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. == MediaWiki Usage Guide == * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:FAQ MediaWiki FAQ] * [https://lists.wikimedia.org/postorius/lists/mediawiki-announce.lists.wikimedia.org/ MediaWiki release mailing list] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language] * [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki] qsxmtuh9rupexyaab07jw6jw86m80xz 341 335 2023-12-14T00:42:36Z Admin 1 341 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care. == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow == need blurb here * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. 5c6m9wm0b9cj77ev0gmvuvyybzmcpns 342 341 2023-12-14T00:43:25Z Admin 1 Admin moved page [[Main Page]] to [[Openwater Wiki]] 341 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care. == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow == need blurb here * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. 5c6m9wm0b9cj77ev0gmvuvyybzmcpns 370 342 2023-12-14T01:37:38Z KedarGrama 6 /* Gen 2 Prototype */ 370 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care. == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow == need blurb here * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. mi1gp16309fpizizf9eg1d8ky0pzkbd 371 370 2023-12-14T06:44:27Z Opw12 8 /* Blood Flow */ 371 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care. == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. 2m4qrnxww5j498sbe9m7ftbeeaz23dj 374 371 2023-12-14T18:19:16Z Opw12 8 /* Blood Flow */ 374 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care. == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. a0i203jyp3lc4gxojj78t1im9xb1wdi 571 374 2023-12-16T05:44:11Z 135.180.195.174 /* Openwater Health - Developing technologies with the potential to revolutionize patient care. */ 571 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == At Openwater Health, we are working on a new generation of portable, wearable medical imaging technology and therapeutic devices that could revolutionize how we deliver care to patients. Our technology might cure cancers, clinical depression and enable rapid identification of large vessel occlusions to decrease the time to intervention in stroke victims, saving lives. Additionally, our therapeutic interventions could benefit patient populations suffering from Alzheimer's Disease and Neuropsychiatric Diseases like Addiction and Anxiety. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. This is being done to create an open source community freely distributing our non-invasive diagnostic and therapeutic technologies. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. qtctaep075s43huu92sjte0zmrglnj6 577 571 2023-12-16T20:51:47Z KedarGrama 6 /* Openwater Health - Developing technologies with the potential to revolutionize patient care */ 577 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of [[Patents]] in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. 1mq1mr6o1kyf7zlwcx88dipy6rv26m6 605 577 2023-12-18T21:52:06Z Soren 11 605 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. lgjcfzjiej49ywmifos2z2som6b6tzm 607 605 2023-12-18T23:19:32Z Admin 1 Protected "[[Openwater Wiki]]" ([Edit=Allow only autoconfirmed users] (indefinite) [Move=Allow only autoconfirmed users] (indefinite)) 605 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. lgjcfzjiej49ywmifos2z2som6b6tzm 612 607 2023-12-19T04:35:00Z Admin 1 Removed protection from "[[Openwater Wiki]]" 605 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. lgjcfzjiej49ywmifos2z2som6b6tzm 613 612 2023-12-19T04:35:32Z 157.131.152.44 613 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. I'm an anonymous test line! 04gr2bjtofi9a5n1maqvuz7fshoreq3 614 613 2023-12-19T04:35:54Z Admin 1 Protected "[[Openwater Wiki]]" ([Edit=Allow only autoconfirmed users] (indefinite) [Move=Allow only autoconfirmed users] (indefinite)) 613 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. I'm an anonymous test line! 04gr2bjtofi9a5n1maqvuz7fshoreq3 615 614 2023-12-19T04:37:38Z Admin 1 615 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. lgjcfzjiej49ywmifos2z2som6b6tzm 619 615 2023-12-19T18:31:17Z Admin 1 619 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. Information about community projects built on Openwater can be found on the [[Community]] page. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. d71v6e05lz99ij6ey0w9ari196nn273 638 619 2023-12-19T18:56:12Z Soren 11 638 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. Information about community projects built on Openwater can be found on the [[Community]] page. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. Our patents can be found [https://drive.google.com/drive/folders/1QcqfGXJawaKrGk_beLsIcPPdyCXTtOCq?usp=sharing here]. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. gkvyt8dxwq62o029mf4pmge2510p9fu 640 638 2023-12-20T00:11:04Z OpenwaterEthan 7 640 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. Information about community projects built on Openwater can be found on the [[Community]] page. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. '''Gen 2 Prototype''' - uses a [[USTX Board|custom designed driver]] and the new [[Neuromodulation|OpenTFUS software]]. '''Gen 1 Prototype''' - uses a verasonics driver and the [https://github.com/OpenwaterHealth/opw_neuromod_sw code described here]. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. Our patents can be found [https://drive.google.com/drive/folders/1QcqfGXJawaKrGk_beLsIcPPdyCXTtOCq?usp=sharing here]. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. 472zuqkh13rmzqi6o11prgje4393l8f 641 640 2023-12-20T00:11:27Z OpenwaterEthan 7 641 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. Information about community projects built on Openwater can be found on the [[Community]] page. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. '''Gen 2 Prototype''' - uses a [[USTX Board|custom designed driver]] and the new [[Neuromodulation|OpenTFUS software]]. '''Gen 1 Prototype''' - uses a Verasonics driver and the [https://github.com/OpenwaterHealth/opw_neuromod_sw code described here]. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. Our patents can be found [https://drive.google.com/drive/folders/1QcqfGXJawaKrGk_beLsIcPPdyCXTtOCq?usp=sharing here]. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. md2api21t37c7phwiinyd5z6q1b13h8 645 641 2023-12-20T00:39:06Z Openwaterpete 5 /* Neuromodulation */ 645 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. Information about community projects built on Openwater can be found on the [[Community]] page. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. '''Gen 2 Prototype''' - uses a [[USTX Board|custom designed driver USTX]] and the new [[Neuromodulation|OpenTFUS software]]. '''Gen 1 Prototype''' - uses a Verasonics driver and the [https://github.com/OpenwaterHealth/opw_neuromod_sw code described here]. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. Our patents can be found [https://drive.google.com/drive/folders/1QcqfGXJawaKrGk_beLsIcPPdyCXTtOCq?usp=sharing here]. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. ht3ggm906swh3394j0h3c9vs7wucnq7 650 645 2023-12-20T16:32:14Z Openwaterpete 5 /* Neuromodulation */ 650 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. Information about community projects built on Openwater can be found on the [[Community]] page. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == [[File:Ustx box photo.png|thumb|figure 2]] The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. '''Gen 2 Prototype''' - uses a [[USTX Board|custom designed driver USTX]] and the new [[Neuromodulation|OpenTFUS software]]. '''Gen 1 Prototype''' - uses a Verasonics driver and the [https://github.com/OpenwaterHealth/opw_neuromod_sw code described here]. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. Our patents can be found [https://drive.google.com/drive/folders/1QcqfGXJawaKrGk_beLsIcPPdyCXTtOCq?usp=sharing here]. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. gnfp2qqzel9ji61nlu9qgk99j78hyfv 651 650 2023-12-20T16:33:15Z Openwaterpete 5 651 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. Information about community projects built on Openwater can be found on the [[Community]] page. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. '''Gen 2 Prototype''' - uses a [[USTX Board|custom designed driver USTX]] and the new [[Neuromodulation|OpenTFUS software]]. '''Gen 1 Prototype''' - uses a Verasonics driver and the [https://github.com/OpenwaterHealth/opw_neuromod_sw code described here]. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. Our patents can be found [https://drive.google.com/drive/folders/1QcqfGXJawaKrGk_beLsIcPPdyCXTtOCq?usp=sharing here]. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. qcpogkh6t05fmsuab2168gunxblki0h 656 651 2023-12-20T22:37:28Z KedarGrama 6 /* Neuromodulation */ 656 wikitext text/x-wiki == Openwater Health - Developing technologies with the potential to revolutionize patient care == Openwater Health is spearheading a revolution in medical technology, developing portable, wearable imaging and therapeutic devices with the potential to dramatically improve patient care. Our focus lies in areas like cancer treatment, stroke diagnosis, and neurological disease management, offering non-invasive solutions that could save lives and enhance well-being. The purpose of this Wiki page is to summarize the entire Openwater technical database and give guidance on the files in the OpenwaterHealth repository. Information about community projects built on Openwater can be found on the [[Community]] page. == Blood Flow == The technology of Openwater Blood Flow devices is a novel method that utilizes light to measure and quantify blood flow within tissue. The technology utilizes highly refined laser light, emitting an extremely narrow range of wavelengths for very brief periods of time. As this specific type of light passes through tissue, its properties can be altered by the flow or movement of blood. Once the light exits the tissue, detecting and quantifying the small changes in flow is achieved by illuminating the transmitted light onto sensors with numerous tiny yet highly sensitive pixels. To accomplish this all, the technique employs simple, yet very sophisticated lasers and sensors that are either already in production for, or under development for, the mass market consumer electronics supply chain. This ultimately allows the device to be scaled down in cost dramatically compared to any other medical device, while providing unique and important biological information. Additional details on this technology and its application to stroke diagnosis can be found below: * [[Whitepaper|Blood Flow Technology and Stroke Diagnosis]] === Gen 2 Prototype === The Blood Flow Gen 2 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. This device is based on the same technologies utilized in the Blood Flow Gen 1 Device, while being heavily optimized for clinical deployment. The device offers real-time and accurate assessment of cerebral blood flow. These devices have been used in clinical trials to demonstrate its strong correlation with comparable clinical standards as well as demonstrate its ability to accurately diagnose large vessel occlusions in stroke patients. Additional details about the device's technology, hardware, and software can be found below: * [[Blood Flow Gen 2 Hardware]] * [[Blood Flow Gen 2 Software Architecture]] * [[Blood Flow Gen 2 Software]] * [[Blood Flow Gen 2 LVO Classification and Analysis]] === Gen 1 Prototype === The Blood Flow Gen 1 Device is a novel medical device designed for the noninvasive measurement of blood flow in the human brain. The device employs both laser and sensor setups never utilized before. This first of its kind device was used for the first clinical studies of the technology as well as for preliminary clinical work detecting strokes. It provided a strong foundation for the development of the [[Blood Flow Gen 2 Hardware|Blood Flow Gen 2 Device]]. Additional details about the device's hardware and software can be found below: * [[Theory of Operation]] * [[Blood Flow Gen 1 Hardware]] * [[Blood Flow Gen 1 Software]] == Neuromodulation == The [[Neuromodulation|Open-TFUS Neuromodulation Platform]] is an ultrasound system that transmits focused ultrasound beams into a subject’s brain with the intention to be able to treat a variety of neurological diseases. The device works by uniquely employing a small array of low-frequency ultrasound transducers in order to precisely steer a small amount of energy to target regions within the brain. This platform has been used in a human clinical trial for depression. '''Gen 2 Prototype''' - uses a [[Openwater Ultrasound Transmit Module (USTX)|custom designed driver USTX]] and the new [[Neuromodulation|OpenTFUS software]]. '''Gen 1 Prototype''' - uses a Verasonics driver and the [https://github.com/OpenwaterHealth/opw_neuromod_sw code described here]. == Oncolysis == The [[Oncolysis|Openwater Preclinical Oncolysis Prototype]] is a therapeutic ultrasound system that uses low-frequency ultrasound with specific acoustic parameters in order to specifically target cancer cells while sparing surrounding healthy tissue. This platform has been used in a preclinical small animal study, demonstrating its performance. == Holographic Acousto-optic Imaging == The Openwater [[Acousto Optic|Holographic Acousto-optic Imaging]] setup is a novel imaging modality employing both optical and ultrasound modalities to provide the functional information of near-infrared spectroscopy and spatial sensitivity of ultrasound. These devices have been used in a variety of pre-clinical studies. == Laboratory Testing == Benchtop performance testing and characterization is a fundamental and critical aspect of developing any medical devices. The following pages highlight some of the most important test setups that are used for optical and ultrasound laboratory testing: * [[Phantoms|Tissue Mimicking Phantoms]] * [[Camera Test Rig|Camera Performance Test Setup]] * [[Laser Characterization]] == Patents == Openwater has filed and been granted an extensive portfolio of patents in sensors, lasers, ultrasound, and more. Our patents can be found [https://drive.google.com/drive/folders/1QcqfGXJawaKrGk_beLsIcPPdyCXTtOCq?usp=sharing here]. == Regulatory == Openwater has engaged a variety of regulatory bodies for its many devices. A comprehensive record of this can be found within the [[Regulatory]] page. == Peer Reviewed Publications == A full listing of manuscripts employing Openwater devices can be found on the [[Publications]] page. a5525h8cnsp8jnhej22gpipl82sq7q8 0 3 3 2023-12-12T20:08:46Z OpenwaterAndrew 3 Created page with "Please visit the opw_phantoms GitHub repository for the complete collection of files referenced on this page. Openwater has developed several phantoms for use with optical, ultrasound, and acousto-optic devices. These phantoms are designed to simulate the required characteristics of human tissue. The phantoms were essential to establish design goals and prove system stability. = Blood Flow System Phantoms = Phantoms used for the blood flow system are designed to simul..." 3 wikitext text/x-wiki Please visit the opw_phantoms GitHub repository for the complete collection of files referenced on this page. Openwater has developed several phantoms for use with optical, ultrasound, and acousto-optic devices. These phantoms are designed to simulate the required characteristics of human tissue. The phantoms were essential to establish design goals and prove system stability. = Blood Flow System Phantoms = Phantoms used for the blood flow system are designed to simulate the optical characteristics of tissue and vasculature under investigation, most importantly scattering and absorption properties. For background on the theory of operation of these devices, please see the Gen 1 system White Paper. The Generation 1 and Generation 2 system pages describe two blood flow systems developed by Openwater. Two types of phantoms are most commonly used Static phantoms. These are composed of a blend of scattering and absorbing powders within a thermoset polymer. They are typically used to collect basic measurements with no motion, often for checking system performance or collecting baseline speckle contrast measurements. The amount of both the scattering and absorbing powders was iterated on to duplicate the throughput optical signal obtained from an average person’s skull. Flow phantoms. These are composed of a similar block of absorptive/scattering material, in which channels are embedded for flowing fluid. This simulates blood flow at different depths beneath the surface of the tissue. == Static Phantom Design and Manufacturing == At a high level, preparing static phantoms involves the following steps: Design and fabrication of the positive mold, which must be matched to the hardware device being tested. The device optodes (source exit window and detector entrance windows) must be in close contact with the surface of the phantom, as they would when performing a measurement on a human subject. Construction of a negative mold in flexible polymer from the positive. Mixing and pouring the scattering/absorbing polymer mixture into the mold. Casting and curing the phantom in the flexible mold. Finishing and polishing operations. See the Gen 2 Phantom Document for material ratios and construction processes. The shape of the phantom must be designed to fit the desired device under test. Optodes, especially the detector entrance windows, must be in close contact with the surface of the phantom. Contact pressure and location must also be repeatable when the device is repeatedly removed and reinstalled on the phantom. See the phantom design folder for detail on two designs built to date: See 7000-0234 and associated files for the design of a phantom and mount for testing a complete headset. See 3000-0725 for a simplified phantom disc design for testing individual sensor modules with straps Static phantom and mount for Gen 2 Blood Flow headset See the tester folder for a manufacturing test assembly used to check optical properties of static phantoms using an off-the-shelf continuous wave laser and photodiode detector, to ensure that optical properties are repeatable when fabricating multiple units. Repeatability testing has been accomplished to assure acceptable levels of measurement variation when performing repeated measurements with a single device. Coefficient of Variation (CoV = standard deviation / mean) is typically used to quantify variation, where each sample is the average contrast value during a scan. For the designs provided, CoV is expected to be <2% for full headset on 7000-0234 headset phantom <2% for simplified sensor module on 3000-0725 disc phantom == Flow Phantom Design and Manufacturing == Characterizing the sensitivity of the Openwater blood flow device to subsurface fluid flow was accomplished with a flow phantom described in the documents referenced below. The fluid flow phantom allowed Openwater to probe both depth and speed of fluid moving below the surface of a scattering and absorbing medium. Flow phantom preparation requires additional steps to create internal channels for fluid flow. Once the desired phantom is made with the inside channels, two tubes are affixed to each depth level needed in the design. A plate is glued to the channels so the scattering liquid can reach the closest adjacent channel. The fluid flow will transverse back and forth on each layer then exit. There is a precision liquid injection system that will move the liquid from a syringe though the phantom at a constant speed. An example of the flow system shown below: Flow phantom test setup The scattering liquid formula is described below: Phantom Construction Instructions. The Gen 2 Phantom Document contains additional detail on flow phantom construction. = Ultrasound System Phantoms = Ultrasound phantoms are essential for the evaluation of ultrasound transducer acoustic output and the tuning required to calibrate a focusing transducer system. Different phantoms are used depending on what is being evaluated. The following document illustrates the composition of Agarose and gelatin with different amounts of alcohol and water to make ultrasound phantoms; the speed of sound within the phantom matrix can be adjusted to meet the experimental criteria: Gelatin and Agarose Phantom Making Instructions. These phantoms do not have the same scattering or attenuation of tissue but they are very useful for quick phantom preparation and coupling of ultrasound energy to various targets including water baths. == IEC acoustical thermal testing == The phantom recipe from the following two papers is recommended in the IEC standard for acoustic testing. The material may be tuned to different acoustic properties (representing different tissues), but in general they have the same material properties of tissue so they can be used for thermal testing where the attenuation of the ultrasound beam is important. For thermal testing meant to validate simulations and also to validate temperature rise for given use cases, these phantoms should be used. From IEC standard for acoustical thermal testing: -reference [1] Thermal phantom recipe for testing of HIFU beams: -reference [2] The reference below describes a long-lasting tissue phantom that can be used for acoustic and thermal testing. It is recommended for use as it was authored by FDA staff: -reference[3] Tissue mimicking tissue phantom for matching acoustic properties are used to evaluate safety on ultrasound therapeutic and diagnostic equipment. This is a detailed recipe that can replace the recipes found in the diagnostic and therapeutic ultrasound safety standards (IEC 60601-2-37). Souza 2016. = Acousto-Optic Phantoms = See page here for an overview of Openwater Acousto-Optic Systems. Design detail on rat and vein phantoms referenced in the acousto-optic wiki can be found here. = References = [1] MADSEN, EL., ZAGZEBSKI, JA. and FRANK, GR. Oil-in-gelatin dispersions for use as ultrasonically tissue-mimicking materials. Ultrasound in Med. & Biol., 1982, 8: 277-287. [2] CHARACTERIZING AN AGAR/GELATIN PHANTOM FOR IMAGE GUIDED DOSING AND FEEDBACK CONTROL OF HIGH-INTENSITY FOCUSED ULTRASOUND [3] A tissue-mimicking material (TMM) for the acoustic and thermal characterization of high-intensity focused ultrasound (HIFU) devices. (authored by FDA staff) bvv64hv44twfob18sxh9qg3xv57pfld 50 3 2023-12-12T22:35:39Z 50.227.118.138 50 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_phantoms opw_phantoms GitHub repository] for the complete collection of files referenced on this page.''' = Blood Flow System Phantoms = Openwater has developed several phantoms for use with optical, ultrasound, and acousto-optic devices. These phantoms simulate the required characteristics of human tissue and have been instrumental in establishing design goals and proving system stability. == Overview == Phantoms used for the blood flow system simulate the optical characteristics of tissue and vasculature under investigation, notably scattering and absorption properties. For background on the theory of operation, refer to the Gen 1 system White Paper. === Types of Phantoms === Two types of phantoms are commonly used: * '''Static Phantoms:''' Composed of a blend of scattering and absorbing powders within a thermoset polymer, primarily for basic measurements without motion. * '''Flow Phantoms:''' Composed of a similar block of absorptive/scattering material with embedded channels for flowing fluid, simulating blood flow at different depths. == Static Phantom Design and Manufacturing == Designing static phantoms involves: * Fabrication of the positive mold, closely matching the hardware device being tested. * Construction of a negative mold in flexible polymer from the positive. * Pouring the scattering/absorbing polymer mixture into the mold. * Casting, curing, finishing, and polishing the phantom. The phantom shape must fit the device under test, ensuring close contact with optodes. Repeatability and pressure must be maintained during device removal and reinstallation. Refer to the Gen 2 Phantom Document for detailed construction processes. == Flow Phantom Design and Manufacturing == Characterizing the sensitivity of the blood flow device to subsurface fluid flow utilized a flow phantom. It allowed probing of fluid depth and speed below the surface of a scattering and absorbing medium. Construction involves creating internal channels for fluid flow, enabling depth and speed exploration. = Ultrasound System Phantoms = Ultrasound phantoms evaluate ultrasound transducer acoustic output and calibration. Different phantoms serve different evaluation purposes. == Gelatin and Agarose Phantoms == These phantoms, made of Agarose and gelatin, adjust the speed of sound within the phantom matrix. They are quick to prepare but lack tissue-like scattering or attenuation. == IEC Acoustical Thermal Testing == The phantom recipe recommended in IEC standards for acoustic testing adjusts to different acoustic properties and is used for thermal testing to validate simulations and temperature rise for specific use cases. = Acousto-Optic Phantoms = See Openwater Acousto-Optic Systems for design details on rat and vein phantoms referenced in the acousto-optic wiki. = References = [1] MADSEN, EL., ZAGZEBSKI, JA. and FRANK, GR. Oil-in-gelatin dispersions for use as ultrasonically tissue-mimicking materials. [2] CHARACTERIZING AN AGAR/GELATIN PHANTOM FOR IMAGE GUIDED DOSING AND FEEDBACK CONTROL OF HIGH-INTENSITY FOCUSED ULTRASOUND [3] A tissue-mimicking material (TMM) for the acoustic and thermal characterization of high-intensity focused ultrasound (HIFU) devices. (authored by FDA staff) 9cvlnyb5p0ekkrym6bwrfayb2keu37v 149 50 2023-12-13T12:45:02Z OpenwaterAndrew 3 149 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_phantoms opw_phantoms GitHub repository] for the complete collection of files referenced on this page.''' = Blood Flow System Phantoms = Openwater has developed several phantoms for use with optical, ultrasound, and acousto-optic devices. These phantoms simulate the required characteristics of human tissue and have been instrumental in establishing design goals and proving system stability. == Overview == Phantoms used for the blood flow system simulate the optical characteristics of tissue and vasculature under investigation, notably scattering and absorption properties. For background on the theory of operation, refer to the Gen 1 system White Paper. === Types of Phantoms === Two types of phantoms are commonly used: * '''Static Phantoms:''' Composed of a blend of scattering and absorbing powders within a thermoset polymer, primarily for basic measurements without motion. * '''Flow Phantoms:''' Composed of a similar block of absorptive/scattering material with embedded channels for flowing fluid, simulating blood flow at different depths. == Static Phantom Design and Manufacturing == Designing static phantoms involves: * Fabrication of the positive mold, closely matching the hardware device being tested. * Construction of a negative mold in flexible polymer from the positive. * Pouring the scattering/absorbing polymer mixture into the mold. * Casting, curing, finishing, and polishing the phantom. The phantom shape must fit the device under test, ensuring close contact with optodes. Repeatability and pressure must be maintained during device removal and reinstallation. Refer to the Gen 2 Phantom Document for detailed construction processes. [[File:Static phantom.PNG|thumb|Static phantom and mount for Gen 2 Blood Flow headset]] == Flow Phantom Design and Manufacturing == Characterizing the sensitivity of the blood flow device to subsurface fluid flow utilized a flow phantom. It allowed probing of fluid depth and speed below the surface of a scattering and absorbing medium. Construction involves creating internal channels for fluid flow, enabling depth and speed exploration. [[File:Flow phantom.PNG|thumb|Flow phantom test setup]] = Ultrasound System Phantoms = Ultrasound phantoms evaluate ultrasound transducer acoustic output and calibration. Different phantoms serve different evaluation purposes. == Gelatin and Agarose Phantoms == These phantoms, made of Agarose and gelatin, adjust the speed of sound within the phantom matrix. They are quick to prepare but lack tissue-like scattering or attenuation. == IEC Acoustical Thermal Testing == The phantom recipe recommended in IEC standards for acoustic testing adjusts to different acoustic properties and is used for thermal testing to validate simulations and temperature rise for specific use cases. = Acousto-Optic Phantoms = See Openwater Acousto-Optic Systems for design details on rat and vein phantoms referenced in the acousto-optic wiki. = References = [1] MADSEN, EL., ZAGZEBSKI, JA. and FRANK, GR. Oil-in-gelatin dispersions for use as ultrasonically tissue-mimicking materials. [2] CHARACTERIZING AN AGAR/GELATIN PHANTOM FOR IMAGE GUIDED DOSING AND FEEDBACK CONTROL OF HIGH-INTENSITY FOCUSED ULTRASOUND [3] A tissue-mimicking material (TMM) for the acoustic and thermal characterization of high-intensity focused ultrasound (HIFU) devices. (authored by FDA staff) 62ui5qwebgf01d1yqw3xyxy7ooo2i58 154 149 2023-12-13T13:15:02Z OpenwaterAndrew 3 154 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_phantoms opw_phantoms GitHub repository] for the complete collection of files referenced on this page.''' Openwater has developed several phantoms for use with optical, ultrasound, and acousto-optic devices. These phantoms simulate the required characteristics of human tissue and have been instrumental in establishing design goals and proving system stability. = Blood Flow System Phantoms = Phantoms used for the blood flow system are designed to simulate the optical characteristics of tissue and vasculature under investigation, most importantly scattering and absorption properties. For background on the theory of operation of these devices, please see the [https://github.com/OpenwaterInternet/opw_bloodflow_gen1_hw/blob/f374203e05bdc9f10b2e045c5fd7239875cdf66c/gen1%20BF%20White%20Paper.pdf Gen 1 system White Paper]. The [https://docs.google.com/document/d/1KmF-bdsAkq_0574uFZ2Ls2jU2u78tlmQNApDsGxCf1U/edit Generation 1] and [https://docs.google.com/document/d/1Q3LOBOVczytu94sE3Mx3TINynO3osimd-jlgie6Kn7M/edit#heading=h.jrx3ehjjesbe Generation 2] system pages describe two blood flow systems developed by Openwater. Two types of phantoms are commonly used: * '''Static Phantoms:''' Composed of a blend of scattering and absorbing powders within a thermoset polymer, primarily for basic measurements without motion. * '''Flow Phantoms:''' Composed of a similar block of absorptive/scattering material with embedded channels for flowing fluid, simulating blood flow at different depths. == Static Phantom Design and Manufacturing == Designing static phantoms involves: * Fabrication of the positive mold, closely matching the hardware device being tested. * Construction of a negative mold in flexible polymer from the positive. * Pouring the scattering/absorbing polymer mixture into the mold. * Casting, curing, finishing, and polishing the phantom. See the [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/optical/D0062%20-%20Gen2%20Optical%20Phantom%20Recipe.pdf Gen 2 Phantom Document] for material ratios and construction processes. The shape of the phantom must be designed to fit the desired device under test. Optodes, especially the detector entrance windows, must be in close contact with the surface of the phantom. Contact pressure and location must also be repeatable when the device is repeatedly removed and reinstalled on the phantom. [[File:Static phantom.PNG|thumb|Static phantom and mount for Gen 2 Blood Flow headset]] See the [https://github.com/OpenwaterInternet/opw_phantoms/tree/main/optical/static_phantoms/phantom_design phantom design folder] for detail on two designs built to date: See 7000-0234 and associated files for the design of a phantom and mount for testing a complete headset. See 3000-0725 for a simplified phantom disc design for testing individual sensor modules with straps See the [https://github.com/OpenwaterInternet/opw_phantoms/tree/main/optical/static_phantoms/tester tester folder] for a manufacturing test assembly used to check optical properties of static phantoms using an off-the-shelf continuous wave laser and photodiode detector, to ensure that optical properties are repeatable when fabricating multiple units. Repeatability testing has been accomplished to assure acceptable levels of measurement variation when performing repeated measurements with a single device. Coefficient of Variation (CoV = standard deviation / mean) is typically used to quantify variation, where each sample is the average contrast value during a scan. For the designs provided, CoV is expected to be <2% for full headset on 7000-0234 headset phantom <2% for simplified sensor module on 3000-0725 disc phantom == Flow Phantom Design and Manufacturing == Characterizing the sensitivity of the Openwater blood flow device to subsurface fluid flow was accomplished with a flow phantom described in the documents referenced below. The fluid flow phantom allowed Openwater to probe both depth and speed of fluid moving below the surface of a scattering and absorbing medium. [[File:Flow phantom.PNG|thumb|Flow phantom test setup]] Flow phantom preparation requires additional steps to create internal channels for fluid flow. Once the desired phantom is made with the inside channels, two tubes are affixed to each depth level needed in the design. A plate is glued to the channels so the scattering liquid can reach the closest adjacent channel. The fluid flow will transverse back and forth on each layer then exit. There is a precision liquid injection system that will move the liquid from a syringe though the phantom at a constant speed. The scattering liquid formula is described in [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/ultrasound%20and%20AO/Phantom%20Construction%20Instructions.pdf Phantom Construction Instructions]. The [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/optical/D0062%20-%20Gen2%20Optical%20Phantom%20Recipe.pdf Gen 2 Phantom Document] contains additional detail on flow phantom construction. = Ultrasound System Phantoms = Ultrasound phantoms are essential for the evaluation of ultrasound transducer acoustic output and the tuning required to calibrate a focusing transducer system. Different phantoms are used depending on what is being evaluated. == Gelatin and Agarose Phantoms == The following document illustrates the composition of Agarose and gelatin with different amounts of alcohol and water to make ultrasound phantoms; the speed of sound within the phantom matrix can be adjusted to meet the experimental criteria: [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/ultrasound%20and%20AO/Gelatin%20and%20agarose%20phantom%20making%20instructions.pdf Gelatin and Agarose Phantom Making Instructions]. These phantoms do not have the same scattering or attenuation of tissue but they are very useful for quick phantom preparation and coupling of ultrasound energy to various targets including water baths. == IEC Acoustical Thermal Testing == The phantom recipe from the following two papers is recommended in the IEC standard for acoustic testing. The material may be tuned to different acoustic properties (representing different tissues), but in general they have the same material properties of tissue so they can be used for thermal testing where the attenuation of the ultrasound beam is important. For thermal testing meant to validate simulations and also to validate temperature rise for given use cases, these phantoms should be used. From IEC standard for acoustical thermal testing: -reference [1] Thermal phantom recipe for testing of HIFU beams: -reference [2] The reference below describes a long-lasting tissue phantom that can be used for acoustic and thermal testing. It is recommended for use as it was authored by FDA staff: -reference[3] Tissue mimicking tissue phantom for matching acoustic properties are used to evaluate safety on ultrasound therapeutic and diagnostic equipment. This is a detailed recipe that can replace the recipes found in the diagnostic and therapeutic ultrasound safety standards (IEC 60601-2-37). Souza 2016. = Acousto-Optic Phantoms = See Openwater Acousto-Optic Systems for an overview of Openwater Acousto-Optic Systems. Design detail on rat and vein phantoms referenced in the acousto-optic wiki can be found [https://github.com/OpenwaterInternet/opw_phantoms/tree/2b92ef5d5744138f787cc84a35ee4329d8e27fbc/ultrasound_and_AO/Rat_and_vein_phantoms here]. = References = [1] MADSEN, EL., ZAGZEBSKI, JA. and FRANK, GR. Oil-in-gelatin dispersions for use as ultrasonically tissue-mimicking materials. [2] CHARACTERIZING AN AGAR/GELATIN PHANTOM FOR IMAGE GUIDED DOSING AND FEEDBACK CONTROL OF HIGH-INTENSITY FOCUSED ULTRASOUND [3] A tissue-mimicking material (TMM) for the acoustic and thermal characterization of high-intensity focused ultrasound (HIFU) devices. (authored by FDA staff) hj1r3riduh0j0pu1632zdctchn36upl 155 154 2023-12-13T13:35:52Z OpenwaterAndrew 3 155 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_phantoms opw_phantoms GitHub repository] for the complete collection of files referenced on this page.''' Openwater has developed several phantoms for use with optical, ultrasound, and acousto-optic devices. These phantoms simulate the required characteristics of human tissue and have been instrumental in establishing design goals and proving system stability. = Blood Flow System Phantoms = Phantoms used for the blood flow system are designed to simulate the optical characteristics of tissue and vasculature under investigation, most importantly scattering and absorption properties. For background on the theory of operation of these devices, please see the [https://github.com/OpenwaterInternet/opw_bloodflow_gen1_hw/blob/f374203e05bdc9f10b2e045c5fd7239875cdf66c/gen1%20BF%20White%20Paper.pdf Gen 1 system White Paper]. The [https://docs.google.com/document/d/1KmF-bdsAkq_0574uFZ2Ls2jU2u78tlmQNApDsGxCf1U/edit Generation 1] and [https://docs.google.com/document/d/1Q3LOBOVczytu94sE3Mx3TINynO3osimd-jlgie6Kn7M/edit#heading=h.jrx3ehjjesbe Generation 2] system pages describe two blood flow systems developed by Openwater. Two types of phantoms are commonly used: * '''Static Phantoms:''' Composed of a blend of scattering and absorbing powders within a thermoset polymer, primarily for basic measurements without motion. * '''Flow Phantoms:''' Composed of a similar block of absorptive/scattering material with embedded channels for flowing fluid, simulating blood flow at different depths. == Static Phantom Design and Manufacturing == Designing static phantoms involves: * Fabrication of the positive mold, closely matching the hardware device being tested. * Construction of a negative mold in flexible polymer from the positive. * Pouring the scattering/absorbing polymer mixture into the mold. * Casting, curing, finishing, and polishing the phantom. See the [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/optical/D0062%20-%20Gen2%20Optical%20Phantom%20Recipe.pdf Gen 2 Phantom Document] for material ratios and construction processes. The shape of the phantom must be designed to fit the desired device under test. Optodes, especially the detector entrance windows, must be in close contact with the surface of the phantom. Contact pressure and location must also be repeatable when the device is repeatedly removed and reinstalled on the phantom. [[File:Static phantom.PNG|thumb|Static phantom and mount for Gen 2 Blood Flow headset]] See the [https://github.com/OpenwaterInternet/opw_phantoms/tree/main/optical/static_phantoms/phantom_design phantom design folder] for detail on two designs built to date: See 7000-0234 and associated files for the design of a phantom and mount for testing a complete headset. See 3000-0725 for a simplified phantom disc design for testing individual sensor modules with straps See the [https://github.com/OpenwaterInternet/opw_phantoms/tree/main/optical/static_phantoms/tester tester folder] for a manufacturing test assembly used to check optical properties of static phantoms using an off-the-shelf continuous wave laser and photodiode detector, to ensure that optical properties are repeatable when fabricating multiple units. Repeatability testing has been accomplished to assure acceptable levels of measurement variation when performing repeated measurements with a single device. Coefficient of Variation (CoV = standard deviation / mean) is typically used to quantify variation, where each sample is the average contrast value during a scan. For the designs provided, CoV is expected to be * <2% for full headset on 7000-0234 headset phantom * <2% for simplified sensor module on 3000-0725 disc phantom == Flow Phantom Design and Manufacturing == Characterizing the sensitivity of the Openwater blood flow device to subsurface fluid flow was accomplished with a flow phantom described in the documents referenced below. The fluid flow phantom allowed Openwater to probe both depth and speed of fluid moving below the surface of a scattering and absorbing medium. [[File:Flow phantom.PNG|thumb|Flow phantom test setup]] Flow phantom preparation requires additional steps to create internal channels for fluid flow. Once the desired phantom is made with the inside channels, two tubes are affixed to each depth level needed in the design. A plate is glued to the channels so the scattering liquid can reach the closest adjacent channel. The fluid flow will transverse back and forth on each layer then exit. There is a precision liquid injection system that will move the liquid from a syringe though the phantom at a constant speed. The scattering liquid formula is described in [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/ultrasound%20and%20AO/Phantom%20Construction%20Instructions.pdf Phantom Construction Instructions]. The [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/optical/D0062%20-%20Gen2%20Optical%20Phantom%20Recipe.pdf Gen 2 Phantom Document] contains additional detail on flow phantom construction. = Ultrasound System Phantoms = Ultrasound phantoms are essential for the evaluation of ultrasound transducer acoustic output and the tuning required to calibrate a focusing transducer system. Different phantoms are used depending on what is being evaluated. == Gelatin and Agarose Phantoms == The following document illustrates the composition of Agarose and gelatin with different amounts of alcohol and water to make ultrasound phantoms; the speed of sound within the phantom matrix can be adjusted to meet the experimental criteria: [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/ultrasound%20and%20AO/Gelatin%20and%20agarose%20phantom%20making%20instructions.pdf Gelatin and Agarose Phantom Making Instructions]. These phantoms do not have the same scattering or attenuation of tissue but they are very useful for quick phantom preparation and coupling of ultrasound energy to various targets including water baths. == IEC Acoustical Thermal Testing == The phantom recipe from the following two papers is recommended in the IEC standard for acoustic testing. The material may be tuned to different acoustic properties (representing different tissues), but in general they have the same material properties of tissue so they can be used for thermal testing where the attenuation of the ultrasound beam is important. For thermal testing meant to validate simulations and also to validate temperature rise for given use cases, these phantoms should be used. From IEC standard for acoustical thermal testing [1] Thermal phantom recipe for testing of HIFU beams [2] The reference below describes a long-lasting tissue phantom that can be used for acoustic and thermal testing. It is recommended for use as it was authored by FDA staff [3] Tissue mimicking tissue phantom for matching acoustic properties are used to evaluate safety on ultrasound therapeutic and diagnostic equipment. This is a detailed recipe that can replace the recipes found in the diagnostic and therapeutic ultrasound safety standards (IEC 60601-2-37). Souza 2016. = Acousto-Optic Phantoms = See Openwater Acousto-Optic Systems for an overview of Openwater Acousto-Optic Systems. Design detail on rat and vein phantoms referenced in the acousto-optic wiki can be found [https://github.com/OpenwaterInternet/opw_phantoms/tree/2b92ef5d5744138f787cc84a35ee4329d8e27fbc/ultrasound_and_AO/Rat_and_vein_phantoms here]. = References = # MADSEN, EL., ZAGZEBSKI, JA. and FRANK, GR. Oil-in-gelatin dispersions for use as ultrasonically tissue-mimicking materials. # CHARACTERIZING AN AGAR/GELATIN PHANTOM FOR IMAGE GUIDED DOSING AND FEEDBACK CONTROL OF HIGH-INTENSITY FOCUSED ULTRASOUND # A tissue-mimicking material (TMM) for the acoustic and thermal characterization of high-intensity focused ultrasound (HIFU) devices. (authored by FDA staff) ri6rgzjurcfq0r5h9w9yx8mxhkr7thw 156 155 2023-12-13T13:42:01Z OpenwaterAndrew 3 156 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_phantoms opw_phantoms GitHub repository] for the complete collection of files referenced on this page.''' Openwater has developed several phantoms for use with optical, ultrasound, and acousto-optic devices. These phantoms simulate the required characteristics of human tissue and have been instrumental in establishing design goals and proving system stability. = Blood Flow System Phantoms = Phantoms used for the blood flow system are designed to simulate the optical characteristics of tissue and vasculature under investigation, most importantly scattering and absorption properties. For background on the theory of operation of these devices, please see the [https://github.com/OpenwaterInternet/opw_bloodflow_gen1_hw/blob/f374203e05bdc9f10b2e045c5fd7239875cdf66c/gen1%20BF%20White%20Paper.pdf Gen 1 system White Paper]. The [https://docs.google.com/document/d/1KmF-bdsAkq_0574uFZ2Ls2jU2u78tlmQNApDsGxCf1U/edit Generation 1] and [https://docs.google.com/document/d/1Q3LOBOVczytu94sE3Mx3TINynO3osimd-jlgie6Kn7M/edit#heading=h.jrx3ehjjesbe Generation 2] system pages describe two blood flow systems developed by Openwater. Two types of phantoms are commonly used: * '''Static Phantoms:''' Composed of a blend of scattering and absorbing powders within a thermoset polymer, primarily for basic measurements without motion. * '''Flow Phantoms:''' Composed of a similar block of absorptive/scattering material with embedded channels for flowing fluid, simulating blood flow at different depths. == Static Phantom Design and Manufacturing == Designing static phantoms involves: * Fabrication of the positive mold, closely matching the hardware device being tested. * Construction of a negative mold in flexible polymer from the positive. * Pouring the scattering/absorbing polymer mixture into the mold. * Casting, curing, finishing, and polishing the phantom. See the [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/optical/D0062%20-%20Gen2%20Optical%20Phantom%20Recipe.pdf Gen 2 Phantom Document] for material ratios and construction processes. The shape of the phantom must be designed to fit the desired device under test. Optodes, especially the detector entrance windows, must be in close contact with the surface of the phantom. Contact pressure and location must also be repeatable when the device is repeatedly removed and reinstalled on the phantom. [[File:Static phantom.PNG|thumb|Static phantom and mount for Gen 2 Blood Flow headset]] See the [https://github.com/OpenwaterInternet/opw_phantoms/tree/main/optical/static_phantoms/phantom_design phantom design folder] for detail on two designs built to date: See 7000-0234 and associated files for the design of a phantom and mount for testing a complete headset. See 3000-0725 for a simplified phantom disc design for testing individual sensor modules with straps See the [https://github.com/OpenwaterInternet/opw_phantoms/tree/main/optical/static_phantoms/tester tester folder] for a manufacturing test assembly used to check optical properties of static phantoms using an off-the-shelf continuous wave laser and photodiode detector, to ensure that optical properties are repeatable when fabricating multiple units. Repeatability testing has been accomplished to assure acceptable levels of measurement variation when performing repeated measurements with a single device. Coefficient of Variation (CoV = standard deviation / mean) is typically used to quantify variation, where each sample is the average contrast value during a scan. For the designs provided, CoV is expected to be * <2% for full headset on 7000-0234 headset phantom * <2% for simplified sensor module on 3000-0725 disc phantom == Flow Phantom Design and Manufacturing == Characterizing the sensitivity of the Openwater blood flow device to subsurface fluid flow was accomplished with a flow phantom described in the documents referenced below. The fluid flow phantom allowed Openwater to probe both depth and speed of fluid moving below the surface of a scattering and absorbing medium. [[File:Flow phantom.PNG|thumb|Flow phantom test setup]] Flow phantom preparation requires additional steps to create internal channels for fluid flow. Once the desired phantom is made with the inside channels, two tubes are affixed to each depth level needed in the design. A plate is glued to the channels so the scattering liquid can reach the closest adjacent channel. The fluid flow will transverse back and forth on each layer then exit. There is a precision liquid injection system that will move the liquid from a syringe though the phantom at a constant speed. The scattering liquid formula is described in [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/ultrasound%20and%20AO/Phantom%20Construction%20Instructions.pdf Phantom Construction Instructions]. The [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/optical/D0062%20-%20Gen2%20Optical%20Phantom%20Recipe.pdf Gen 2 Phantom Document] contains additional detail on flow phantom construction. = Ultrasound System Phantoms = Ultrasound phantoms are essential for the evaluation of ultrasound transducer acoustic output and the tuning required to calibrate a focusing transducer system. Different phantoms are used depending on what is being evaluated. == Gelatin and Agarose Phantoms == The following document illustrates the composition of Agarose and gelatin with different amounts of alcohol and water to make ultrasound phantoms; the speed of sound within the phantom matrix can be adjusted to meet the experimental criteria: [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/ultrasound%20and%20AO/Gelatin%20and%20agarose%20phantom%20making%20instructions.pdf Gelatin and Agarose Phantom Making Instructions]. These phantoms do not have the same scattering or attenuation of tissue but they are very useful for quick phantom preparation and coupling of ultrasound energy to various targets including water baths. == IEC Acoustical Thermal Testing == The phantom recipe from the following two papers is recommended in the IEC standard for acoustic testing. The material may be tuned to different acoustic properties (representing different tissues), but in general they have the same material properties of tissue so they can be used for thermal testing where the attenuation of the ultrasound beam is important. For thermal testing meant to validate simulations and also to validate temperature rise for given use cases, these phantoms should be used. From IEC standard for acoustical thermal testing [1] Thermal phantom recipe for testing of HIFU beams [2] The reference below describes a long-lasting tissue phantom that can be used for acoustic and thermal testing. It is recommended for use as it was authored by FDA staff [3] Tissue mimicking tissue phantom for matching acoustic properties are used to evaluate safety on ultrasound therapeutic and diagnostic equipment. This is a detailed recipe that can replace the recipes found in the diagnostic and therapeutic ultrasound safety standards (IEC 60601-2-37). Souza 2016. = Acousto-Optic Phantoms = See Openwater Acousto-Optic Systems for an overview of Openwater Acousto-Optic Systems. Design detail on rat and vein phantoms referenced in the acousto-optic wiki can be found [https://github.com/OpenwaterInternet/opw_phantoms/tree/2b92ef5d5744138f787cc84a35ee4329d8e27fbc/ultrasound_and_AO/Rat_and_vein_phantoms here]. = References = # MADSEN, EL., ZAGZEBSKI, JA. and FRANK, GR. [https://www.umbjournal.org/article/0301-5629(82)90034-5/pdf Oil-in-gelatin dispersions for use as ultrasonically tissue-mimicking materials]. Ultrasound in Med. & Biol., 1982, 8: 277-287. # [https://www.umbjournal.org/article/S0301-5629(12)00582-0/fulltext CHARACTERIZING AN AGAR/GELATIN PHANTOM FOR IMAGE GUIDED DOSING AND FEEDBACK CONTROL OF HIGH-INTENSITY FOCUSED ULTRASOUND] # [https://ieeexplore.ieee.org/abstract/document/5953995 A tissue-mimicking material (TMM) for the acoustic and thermal characterization of high-intensity focused ultrasound (HIFU) devices]. (authored by FDA staff) idqeb6t1ufz8knignscm492dna42kuy 157 156 2023-12-13T13:42:13Z OpenwaterAndrew 3 /* References */ 157 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_phantoms opw_phantoms GitHub repository] for the complete collection of files referenced on this page.''' Openwater has developed several phantoms for use with optical, ultrasound, and acousto-optic devices. These phantoms simulate the required characteristics of human tissue and have been instrumental in establishing design goals and proving system stability. = Blood Flow System Phantoms = Phantoms used for the blood flow system are designed to simulate the optical characteristics of tissue and vasculature under investigation, most importantly scattering and absorption properties. For background on the theory of operation of these devices, please see the [https://github.com/OpenwaterInternet/opw_bloodflow_gen1_hw/blob/f374203e05bdc9f10b2e045c5fd7239875cdf66c/gen1%20BF%20White%20Paper.pdf Gen 1 system White Paper]. The [https://docs.google.com/document/d/1KmF-bdsAkq_0574uFZ2Ls2jU2u78tlmQNApDsGxCf1U/edit Generation 1] and [https://docs.google.com/document/d/1Q3LOBOVczytu94sE3Mx3TINynO3osimd-jlgie6Kn7M/edit#heading=h.jrx3ehjjesbe Generation 2] system pages describe two blood flow systems developed by Openwater. Two types of phantoms are commonly used: * '''Static Phantoms:''' Composed of a blend of scattering and absorbing powders within a thermoset polymer, primarily for basic measurements without motion. * '''Flow Phantoms:''' Composed of a similar block of absorptive/scattering material with embedded channels for flowing fluid, simulating blood flow at different depths. == Static Phantom Design and Manufacturing == Designing static phantoms involves: * Fabrication of the positive mold, closely matching the hardware device being tested. * Construction of a negative mold in flexible polymer from the positive. * Pouring the scattering/absorbing polymer mixture into the mold. * Casting, curing, finishing, and polishing the phantom. See the [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/optical/D0062%20-%20Gen2%20Optical%20Phantom%20Recipe.pdf Gen 2 Phantom Document] for material ratios and construction processes. The shape of the phantom must be designed to fit the desired device under test. Optodes, especially the detector entrance windows, must be in close contact with the surface of the phantom. Contact pressure and location must also be repeatable when the device is repeatedly removed and reinstalled on the phantom. [[File:Static phantom.PNG|thumb|Static phantom and mount for Gen 2 Blood Flow headset]] See the [https://github.com/OpenwaterInternet/opw_phantoms/tree/main/optical/static_phantoms/phantom_design phantom design folder] for detail on two designs built to date: See 7000-0234 and associated files for the design of a phantom and mount for testing a complete headset. See 3000-0725 for a simplified phantom disc design for testing individual sensor modules with straps See the [https://github.com/OpenwaterInternet/opw_phantoms/tree/main/optical/static_phantoms/tester tester folder] for a manufacturing test assembly used to check optical properties of static phantoms using an off-the-shelf continuous wave laser and photodiode detector, to ensure that optical properties are repeatable when fabricating multiple units. Repeatability testing has been accomplished to assure acceptable levels of measurement variation when performing repeated measurements with a single device. Coefficient of Variation (CoV = standard deviation / mean) is typically used to quantify variation, where each sample is the average contrast value during a scan. For the designs provided, CoV is expected to be * <2% for full headset on 7000-0234 headset phantom * <2% for simplified sensor module on 3000-0725 disc phantom == Flow Phantom Design and Manufacturing == Characterizing the sensitivity of the Openwater blood flow device to subsurface fluid flow was accomplished with a flow phantom described in the documents referenced below. The fluid flow phantom allowed Openwater to probe both depth and speed of fluid moving below the surface of a scattering and absorbing medium. [[File:Flow phantom.PNG|thumb|Flow phantom test setup]] Flow phantom preparation requires additional steps to create internal channels for fluid flow. Once the desired phantom is made with the inside channels, two tubes are affixed to each depth level needed in the design. A plate is glued to the channels so the scattering liquid can reach the closest adjacent channel. The fluid flow will transverse back and forth on each layer then exit. There is a precision liquid injection system that will move the liquid from a syringe though the phantom at a constant speed. The scattering liquid formula is described in [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/ultrasound%20and%20AO/Phantom%20Construction%20Instructions.pdf Phantom Construction Instructions]. The [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/optical/D0062%20-%20Gen2%20Optical%20Phantom%20Recipe.pdf Gen 2 Phantom Document] contains additional detail on flow phantom construction. = Ultrasound System Phantoms = Ultrasound phantoms are essential for the evaluation of ultrasound transducer acoustic output and the tuning required to calibrate a focusing transducer system. Different phantoms are used depending on what is being evaluated. == Gelatin and Agarose Phantoms == The following document illustrates the composition of Agarose and gelatin with different amounts of alcohol and water to make ultrasound phantoms; the speed of sound within the phantom matrix can be adjusted to meet the experimental criteria: [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/ultrasound%20and%20AO/Gelatin%20and%20agarose%20phantom%20making%20instructions.pdf Gelatin and Agarose Phantom Making Instructions]. These phantoms do not have the same scattering or attenuation of tissue but they are very useful for quick phantom preparation and coupling of ultrasound energy to various targets including water baths. == IEC Acoustical Thermal Testing == The phantom recipe from the following two papers is recommended in the IEC standard for acoustic testing. The material may be tuned to different acoustic properties (representing different tissues), but in general they have the same material properties of tissue so they can be used for thermal testing where the attenuation of the ultrasound beam is important. For thermal testing meant to validate simulations and also to validate temperature rise for given use cases, these phantoms should be used. From IEC standard for acoustical thermal testing [1] Thermal phantom recipe for testing of HIFU beams [2] The reference below describes a long-lasting tissue phantom that can be used for acoustic and thermal testing. It is recommended for use as it was authored by FDA staff [3] Tissue mimicking tissue phantom for matching acoustic properties are used to evaluate safety on ultrasound therapeutic and diagnostic equipment. This is a detailed recipe that can replace the recipes found in the diagnostic and therapeutic ultrasound safety standards (IEC 60601-2-37). Souza 2016. = Acousto-Optic Phantoms = See Openwater Acousto-Optic Systems for an overview of Openwater Acousto-Optic Systems. Design detail on rat and vein phantoms referenced in the acousto-optic wiki can be found [https://github.com/OpenwaterInternet/opw_phantoms/tree/2b92ef5d5744138f787cc84a35ee4329d8e27fbc/ultrasound_and_AO/Rat_and_vein_phantoms here]. = References = # MADSEN, EL., ZAGZEBSKI, JA. and FRANK, GR. [https://www.umbjournal.org/article/0301-5629(82)90034-5/pdf Oil-in-gelatin dispersions for use as ultrasonically tissue-mimicking materials]. Ultrasound in Med. & Biol., 1982, 8:277-287. # [https://www.umbjournal.org/article/S0301-5629(12)00582-0/fulltext CHARACTERIZING AN AGAR/GELATIN PHANTOM FOR IMAGE GUIDED DOSING AND FEEDBACK CONTROL OF HIGH-INTENSITY FOCUSED ULTRASOUND] # [https://ieeexplore.ieee.org/abstract/document/5953995 A tissue-mimicking material (TMM) for the acoustic and thermal characterization of high-intensity focused ultrasound (HIFU) devices]. (authored by FDA staff) fekq96wzzolbg6npg03bn4bbtceq1gv 634 157 2023-12-19T18:43:16Z Admin 1 Protected "[[Phantoms]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 157 wikitext text/x-wiki '''Please visit the [https://github.com/OpenwaterHealth/opw_phantoms opw_phantoms GitHub repository] for the complete collection of files referenced on this page.''' Openwater has developed several phantoms for use with optical, ultrasound, and acousto-optic devices. These phantoms simulate the required characteristics of human tissue and have been instrumental in establishing design goals and proving system stability. = Blood Flow System Phantoms = Phantoms used for the blood flow system are designed to simulate the optical characteristics of tissue and vasculature under investigation, most importantly scattering and absorption properties. For background on the theory of operation of these devices, please see the [https://github.com/OpenwaterInternet/opw_bloodflow_gen1_hw/blob/f374203e05bdc9f10b2e045c5fd7239875cdf66c/gen1%20BF%20White%20Paper.pdf Gen 1 system White Paper]. The [https://docs.google.com/document/d/1KmF-bdsAkq_0574uFZ2Ls2jU2u78tlmQNApDsGxCf1U/edit Generation 1] and [https://docs.google.com/document/d/1Q3LOBOVczytu94sE3Mx3TINynO3osimd-jlgie6Kn7M/edit#heading=h.jrx3ehjjesbe Generation 2] system pages describe two blood flow systems developed by Openwater. Two types of phantoms are commonly used: * '''Static Phantoms:''' Composed of a blend of scattering and absorbing powders within a thermoset polymer, primarily for basic measurements without motion. * '''Flow Phantoms:''' Composed of a similar block of absorptive/scattering material with embedded channels for flowing fluid, simulating blood flow at different depths. == Static Phantom Design and Manufacturing == Designing static phantoms involves: * Fabrication of the positive mold, closely matching the hardware device being tested. * Construction of a negative mold in flexible polymer from the positive. * Pouring the scattering/absorbing polymer mixture into the mold. * Casting, curing, finishing, and polishing the phantom. See the [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/optical/D0062%20-%20Gen2%20Optical%20Phantom%20Recipe.pdf Gen 2 Phantom Document] for material ratios and construction processes. The shape of the phantom must be designed to fit the desired device under test. Optodes, especially the detector entrance windows, must be in close contact with the surface of the phantom. Contact pressure and location must also be repeatable when the device is repeatedly removed and reinstalled on the phantom. [[File:Static phantom.PNG|thumb|Static phantom and mount for Gen 2 Blood Flow headset]] See the [https://github.com/OpenwaterInternet/opw_phantoms/tree/main/optical/static_phantoms/phantom_design phantom design folder] for detail on two designs built to date: See 7000-0234 and associated files for the design of a phantom and mount for testing a complete headset. See 3000-0725 for a simplified phantom disc design for testing individual sensor modules with straps See the [https://github.com/OpenwaterInternet/opw_phantoms/tree/main/optical/static_phantoms/tester tester folder] for a manufacturing test assembly used to check optical properties of static phantoms using an off-the-shelf continuous wave laser and photodiode detector, to ensure that optical properties are repeatable when fabricating multiple units. Repeatability testing has been accomplished to assure acceptable levels of measurement variation when performing repeated measurements with a single device. Coefficient of Variation (CoV = standard deviation / mean) is typically used to quantify variation, where each sample is the average contrast value during a scan. For the designs provided, CoV is expected to be * <2% for full headset on 7000-0234 headset phantom * <2% for simplified sensor module on 3000-0725 disc phantom == Flow Phantom Design and Manufacturing == Characterizing the sensitivity of the Openwater blood flow device to subsurface fluid flow was accomplished with a flow phantom described in the documents referenced below. The fluid flow phantom allowed Openwater to probe both depth and speed of fluid moving below the surface of a scattering and absorbing medium. [[File:Flow phantom.PNG|thumb|Flow phantom test setup]] Flow phantom preparation requires additional steps to create internal channels for fluid flow. Once the desired phantom is made with the inside channels, two tubes are affixed to each depth level needed in the design. A plate is glued to the channels so the scattering liquid can reach the closest adjacent channel. The fluid flow will transverse back and forth on each layer then exit. There is a precision liquid injection system that will move the liquid from a syringe though the phantom at a constant speed. The scattering liquid formula is described in [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/ultrasound%20and%20AO/Phantom%20Construction%20Instructions.pdf Phantom Construction Instructions]. The [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/optical/D0062%20-%20Gen2%20Optical%20Phantom%20Recipe.pdf Gen 2 Phantom Document] contains additional detail on flow phantom construction. = Ultrasound System Phantoms = Ultrasound phantoms are essential for the evaluation of ultrasound transducer acoustic output and the tuning required to calibrate a focusing transducer system. Different phantoms are used depending on what is being evaluated. == Gelatin and Agarose Phantoms == The following document illustrates the composition of Agarose and gelatin with different amounts of alcohol and water to make ultrasound phantoms; the speed of sound within the phantom matrix can be adjusted to meet the experimental criteria: [https://github.com/OpenwaterInternet/opw_phantoms/blob/7aa956a693cb36ca7434037c67c698c352c5c0de/ultrasound%20and%20AO/Gelatin%20and%20agarose%20phantom%20making%20instructions.pdf Gelatin and Agarose Phantom Making Instructions]. These phantoms do not have the same scattering or attenuation of tissue but they are very useful for quick phantom preparation and coupling of ultrasound energy to various targets including water baths. == IEC Acoustical Thermal Testing == The phantom recipe from the following two papers is recommended in the IEC standard for acoustic testing. The material may be tuned to different acoustic properties (representing different tissues), but in general they have the same material properties of tissue so they can be used for thermal testing where the attenuation of the ultrasound beam is important. For thermal testing meant to validate simulations and also to validate temperature rise for given use cases, these phantoms should be used. From IEC standard for acoustical thermal testing [1] Thermal phantom recipe for testing of HIFU beams [2] The reference below describes a long-lasting tissue phantom that can be used for acoustic and thermal testing. It is recommended for use as it was authored by FDA staff [3] Tissue mimicking tissue phantom for matching acoustic properties are used to evaluate safety on ultrasound therapeutic and diagnostic equipment. This is a detailed recipe that can replace the recipes found in the diagnostic and therapeutic ultrasound safety standards (IEC 60601-2-37). Souza 2016. = Acousto-Optic Phantoms = See Openwater Acousto-Optic Systems for an overview of Openwater Acousto-Optic Systems. Design detail on rat and vein phantoms referenced in the acousto-optic wiki can be found [https://github.com/OpenwaterInternet/opw_phantoms/tree/2b92ef5d5744138f787cc84a35ee4329d8e27fbc/ultrasound_and_AO/Rat_and_vein_phantoms here]. = References = # MADSEN, EL., ZAGZEBSKI, JA. and FRANK, GR. [https://www.umbjournal.org/article/0301-5629(82)90034-5/pdf Oil-in-gelatin dispersions for use as ultrasonically tissue-mimicking materials]. Ultrasound in Med. & Biol., 1982, 8:277-287. # [https://www.umbjournal.org/article/S0301-5629(12)00582-0/fulltext CHARACTERIZING AN AGAR/GELATIN PHANTOM FOR IMAGE GUIDED DOSING AND FEEDBACK CONTROL OF HIGH-INTENSITY FOCUSED ULTRASOUND] # [https://ieeexplore.ieee.org/abstract/document/5953995 A tissue-mimicking material (TMM) for the acoustic and thermal characterization of high-intensity focused ultrasound (HIFU) devices]. (authored by FDA staff) fekq96wzzolbg6npg03bn4bbtceq1gv 0 61 251 2023-12-13T20:21:25Z Opw12 8 List of peer reviewed publications using Openwater technologies 251 wikitext text/x-wiki The following is a listing of peer reviewed manuscripts using Openwater technologies: - Gen 2 Blood Flow device correlation with transcranial doppler (TCD) ultrasound - - Link to Medrxiv - Using the Gen 2 Blood Flow device to detect ischemic strokes (Large Vessel Occlusion) - - Link to Medrxiv b46kbs3fj1ran25bq05zsr5ymdtl733 252 251 2023-12-13T20:30:05Z Opw12 8 252 wikitext text/x-wiki The following is a listing of peer reviewed manuscripts using Openwater technologies: * [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 '''Validation of the Openwater wearable optical system: cerebral hemodynamic monitoring during a breath hold maneuver'''] (submitted, under review) ** This manuscript demonstrates strong correlation between the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] and transcranial doppler ultrasound, a clinical gold standard for measuring cerebral blood flow. * '''Working title of the LVO manuscript here''' (in preparation) ** This manuscript demonstrates the ability of the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] and transcranial doppler ultrasound, a clinical gold standard for measuring cerebral blood flow. n26mi2r7vxekgf3ybmayefway4jz3rp 330 252 2023-12-14T00:06:57Z 50.227.118.138 330 wikitext text/x-wiki The following is a listing of peer reviewed manuscripts using Openwater technologies: * [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 '''Validation of the Openwater wearable optical system: cerebral hemodynamic monitoring during a breath hold maneuver'''] (submitted, under review) ** This manuscript demonstrates strong correlation between the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] and transcranial doppler ultrasound, a clinical gold standard for measuring cerebral blood flow. * '''Portable Cerebral Blood Flow Monitor to Detect Large Vessel Occlusion in Suspected Stroke Patients''' (in preparation) ** This manuscript demonstrates the ability of the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] to detect large vessel occlusion in suspected stroke patients. * '''A wearable, steerable, transcranial low-intensity focused ultrasound system''' (in preperation) ** This manuscript describes the first wearable transcranial therapeutic ultrasound [[Neuromodulation|prototype]] using a steerable phased array. ss2mm7jk2h0nu2c1ia8vaah0ipg53k0 508 330 2023-12-15T21:41:06Z KedarGrama 6 508 wikitext text/x-wiki The following is a listing of peer reviewed manuscripts using Openwater technologies: * [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 '''Validation of the Openwater wearable optical system: cerebral hemodynamic monitoring during a breath hold maneuver'''] (submitted, under review) ** This manuscript demonstrates strong correlation between the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] and transcranial doppler ultrasound, a clinical gold standard for measuring cerebral blood flow. * '''[https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 Portable Cerebral Blood Flow Monitor to Detect Large Vessel Occlusion in Suspected Stroke Patients]''' (submitted, under review) ** This manuscript demonstrates the ability of the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] to detect large vessel occlusion in suspected stroke patients. * '''A wearable, steerable, transcranial low-intensity focused ultrasound system''' (in preperation) ** This manuscript describes the first wearable transcranial therapeutic ultrasound [[Neuromodulation|prototype]] using a steerable phased array. 96tvwximm4f328mu11kzy8ruv1cufab 635 508 2023-12-19T18:43:26Z Admin 1 Protected "[[Publications]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 508 wikitext text/x-wiki The following is a listing of peer reviewed manuscripts using Openwater technologies: * [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 '''Validation of the Openwater wearable optical system: cerebral hemodynamic monitoring during a breath hold maneuver'''] (submitted, under review) ** This manuscript demonstrates strong correlation between the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] and transcranial doppler ultrasound, a clinical gold standard for measuring cerebral blood flow. * '''[https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 Portable Cerebral Blood Flow Monitor to Detect Large Vessel Occlusion in Suspected Stroke Patients]''' (submitted, under review) ** This manuscript demonstrates the ability of the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] to detect large vessel occlusion in suspected stroke patients. * '''A wearable, steerable, transcranial low-intensity focused ultrasound system''' (in preperation) ** This manuscript describes the first wearable transcranial therapeutic ultrasound [[Neuromodulation|prototype]] using a steerable phased array. 96tvwximm4f328mu11kzy8ruv1cufab 657 635 2023-12-20T23:35:30Z Soren 11 657 wikitext text/x-wiki The following is a listing of peer reviewed manuscripts using Openwater technologies: * [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 '''Validation of the Openwater wearable optical system: cerebral hemodynamic monitoring during a breath hold maneuver'''] (submitted, under review) ** This manuscript demonstrates strong correlation between the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] and transcranial doppler ultrasound, a clinical gold standard for measuring cerebral blood flow. * '''[https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 Portable Cerebral Blood Flow Monitor to Detect Large Vessel Occlusion in Suspected Stroke Patients]''' (submitted, under review) ** This manuscript demonstrates the ability of the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] to detect large vessel occlusion in suspected stroke patients. * '''A wearable, steerable, transcranial low-intensity focused ultrasound [https://github.com/OpenwaterHealth/opw_neuromod_hw/blob/main/Neuromod_System_Paper_Preprint.pdf system]''' (submitted, under review) ** This manuscript describes the first wearable transcranial therapeutic ultrasound [[Neuromodulation|prototype]] using a steerable phased array. eitgxpxgeda35v89f4low0vjrt0671u 663 657 2023-12-22T19:41:23Z KedarGrama 6 663 wikitext text/x-wiki The following is a listing of peer reviewed manuscripts using Openwater technologies: * [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 '''Validation of the Openwater wearable optical system: cerebral hemodynamic monitoring during a breath hold maneuver'''] (submitted, under review) ** This manuscript demonstrates strong correlation between the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] and transcranial doppler ultrasound, a clinical gold standard for measuring cerebral blood flow. * '''[https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 Portable Cerebral Blood Flow Monitor to Detect Large Vessel Occlusion in Suspected Stroke Patients]''' (submitted, under review) ** This manuscript demonstrates the ability of the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] to detect large vessel occlusion in suspected stroke patients. * '''[https://github.com/OpenwaterHealth/opw_neuromod_hw/blob/main/Neuromod_System_Paper_Preprint.pdf A wearable, steerable, transcranial low-intensity focused ultrasound system]''' (submitted, under review) ** This manuscript describes the first wearable transcranial therapeutic ultrasound [[Neuromodulation|prototype]] using a steerable phased array. e6l0p669cyccbgxakym7h6ugbnlg3dk 668 663 2024-01-02T23:22:10Z KedarGrama 6 668 wikitext text/x-wiki The following is a listing of peer reviewed manuscripts using Openwater technologies: * [https://www.medrxiv.org/content/10.1101/2023.10.11.23296612v1 '''Validation of the Openwater wearable optical system: cerebral hemodynamic monitoring during a breath hold maneuver'''] (submitted, under review) ** This manuscript demonstrates strong correlation between the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] and transcranial doppler ultrasound, a clinical gold standard for measuring cerebral blood flow. * '''[https://www.medrxiv.org/content/10.1101/2023.12.14.23299992v1 Portable Cerebral Blood Flow Monitor to Detect Large Vessel Occlusion in Suspected Stroke Patients]''' (submitted, under review) ** This manuscript demonstrates the ability of the [[Blood Flow Gen 2 Hardware|Gen 2 Blood Flow device]] to detect large vessel occlusion in suspected stroke patients. * '''[https://www.medrxiv.org/content/10.1101/2023.12.22.23300243v1 A wearable, steerable, transcranial low-intensity focused ultrasound system]''' (submitted, under review) ** This manuscript describes the first wearable transcranial therapeutic ultrasound [[Neuromodulation|prototype]] using a steerable phased array. gdb39a6kc3t0zr25eg7xlfzb1auwwju 0 62 256 2023-12-13T20:36:20Z Opw12 8 Overview of Openwater's interactions with regulatory bodies 256 wikitext text/x-wiki Overview of Openwater's interactions with regulatory bodies eupe1w2dhen04hef3h9grrkrd80t3qd 257 256 2023-12-13T20:38:47Z Opw12 8 257 wikitext text/x-wiki Overview of Openwater's interactions with regulatory bodies All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. t8br7lt245bb1r8iuh3xgp3neaedfz3 384 257 2023-12-14T20:46:58Z KedarGrama 6 Initial move of regulatory writeup from google docs to wiki 384 wikitext text/x-wiki All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. '''Openwater LVO Stroke Alert Regulatory''' These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. '''Background''' According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to reperfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. '''Communication With the FDA''' * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ljby4naj6if6hfu7htzw7rx7cwcgook 385 384 2023-12-14T20:47:34Z KedarGrama 6 385 wikitext text/x-wiki All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ===Openwater LVO Stroke Alert Regulatory=== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to reperfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. '''Communication With the FDA''' * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] hft66aplu95ts9ofrva5ngb2cso8a8v 386 385 2023-12-14T20:48:30Z KedarGrama 6 /* Background */ 386 wikitext text/x-wiki All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ===Openwater LVO Stroke Alert Regulatory=== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ====Communication With the FDA==== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ciof7fobwmb22yzwu2xcfes0ropl6bb 387 386 2023-12-14T20:52:28Z KedarGrama 6 387 wikitext text/x-wiki ==Introduction== <Please fill a couple of sentences to introduce our products and the regulatory path>. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] eiesm2maz1a8cpdtcmcu1ch78rjilmk 390 387 2023-12-14T21:20:14Z KedarGrama 6 390 wikitext text/x-wiki ==Introduction== <Please fill a couple of sentences to introduce our products and the regulatory path>. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, ● Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi). ● Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb). trbmaoadq5tmk9970vaq55i8f2c7qif 396 390 2023-12-14T22:24:34Z 104.153.246.86 Added introduction 396 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This Wikipedia page is dedicated to providing transparent insight into the regulatory strategy and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA). All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, ● Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi). ● Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb). pk3rjytnbdg2eien51jo6sribczjxr9 397 396 2023-12-14T22:37:50Z 104.153.246.86 Updated Introduction 397 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wikipedia page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, ● Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi). ● Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb). ek9o2fh0z15tsno3h07288wcpfp0z8a 399 397 2023-12-14T22:52:18Z KedarGrama 6 399 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wikipedia page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb). 9lgwkiu1w9r9fj3riyxjkdzfcwfqtiy 400 399 2023-12-14T22:57:47Z Soren 11 400 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wikipedia page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb). === Communication with the FDA === kxn7cjd3kkv2tmj6kw7b405ja5st3ja 401 400 2023-12-14T23:00:28Z KedarGrama 6 401 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wikipedia page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == '''Openwater Low Intensity Focused Ultrasound Therapy System (LOFU)''' == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. === '''Background''' === Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. c9lcv1da1qu425qmag0ytvxt3ap67wx 402 401 2023-12-14T23:04:04Z Soren 11 402 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wikipedia page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. === '''Background''' === Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. hv0arc3fc7vd4gzsxpjs8db2i7kmryb 403 402 2023-12-14T23:04:49Z KedarGrama 6 403 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wikipedia page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. === '''Background''' === Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] 5jah6iav5i3kxdrfszwb30ucldcsgys 404 403 2023-12-14T23:11:48Z Soren 11 404 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. === '''Background''' === Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] hx9syshdfizhg24uv9pvj6lmk7tiff1 405 404 2023-12-14T23:12:41Z Soren 11 405 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] === References === ????????????????? ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. === '''Background''' === Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] g740od7ax3yvm6eklyzryo0ofbd19qf 406 405 2023-12-14T23:14:44Z Soren 11 406 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] === References === ????????????????? ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. === '''Background''' === Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] === References === ??????????????????? mljsnb6f4ijkqrxvkhem315pkqogt7d 506 406 2023-12-15T21:38:26Z KedarGrama 6 506 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. === '''Background''' === Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] hx9syshdfizhg24uv9pvj6lmk7tiff1 507 506 2023-12-15T21:40:10Z KedarGrama 6 507 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. === '''Background''' === Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==References== 1r9cf32hkis5lccmitrvhgt1dgs37uc 556 507 2023-12-16T00:04:51Z KedarGrama 6 /* Background */ 556 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS) (Virani et al., 2020). Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. ===Background=== Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==References== ly3fvhi2sovng5f4hjwg9difrq0yt48 558 556 2023-12-16T00:10:28Z KedarGrama 6 /* Openwater LVO Stroke Alert */ 558 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)<ref>Virani, S. S., Alonso, A., Benjamin, E. J., Bittencourt, M. S., Callaway, C. W., Carson, A. P., Chamberlain, A. M., Chang, A. R., Cheng, S., Delling, F. N., Djousse, L., Elkind, M. S. V., Ferguson, J. F., Fornage, M., Khan, S. S., Kissela, B. M., Knutson, K. L., Kwan, T. W., Lackland, D. T., ... Subcommittee, O. behalf of the A. H. A. C. on E. and P. S. C. and S. S. (2020). Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association. ''Circulation'', ''141'', E139–E596.</ref>. Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020). Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA) (Rennert et al., 2019). Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone (McCarthy et al., 2019). Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state (McCarthy et al., 2019). However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. ===Background=== Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==References== bzoej8vu9b1a2egafgb3smkhapk7fey 560 558 2023-12-16T00:13:02Z KedarGrama 6 /* Openwater LVO Stroke Alert */ 560 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)<ref>Virani, S. S., Alonso, A., Benjamin, E. J., Bittencourt, M. S., Callaway, C. W., Carson, A. P., Chamberlain, A. M., Chang, A. R., Cheng, S., Delling, F. N., Djousse, L., Elkind, M. S. V., Ferguson, J. F., Fornage, M., Khan, S. S., Kissela, B. M., Knutson, K. L., Kwan, T. W., Lackland, D. T., ... Subcommittee, O. behalf of the A. H. A. C. on E. and P. S. C. and S. S. (2020). Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association. ''Circulation'', ''141'', E139–E596.</ref>. Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes (Silva & Nogueira, 2020)<ref>Silva, G. S., & Nogueira, R. G. (2020). Endovascular treatment of acute ischemic stroke. ''CONTINUUM: Lifelong Learning in Neurology'', ''26''(2), 310–331.</ref>. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA)<ref>Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. ''Neurosurgery'', ''85''(suppl_1), S4–S8.</ref>. Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone<ref name=":0">McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In ''Scientific World Journal'' (Vol. 2019).</ref>. Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state<ref name=":0" />. However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit (Goyal et al., 2016). Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred (Goyal et al., 2016). The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system (Kunz et al., 2020). These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. ===Background=== Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==References== 7y4f1s35l4lydjx79cv15uwd6pos510 562 560 2023-12-16T00:19:12Z KedarGrama 6 562 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)<ref>Virani, S. S., Alonso, A., Benjamin, E. J., Bittencourt, M. S., Callaway, C. W., Carson, A. P., Chamberlain, A. M., Chang, A. R., Cheng, S., Delling, F. N., Djousse, L., Elkind, M. S. V., Ferguson, J. F., Fornage, M., Khan, S. S., Kissela, B. M., Knutson, K. L., Kwan, T. W., Lackland, D. T., ... Subcommittee, O. behalf of the A. H. A. C. on E. and P. S. C. and S. S. (2020). Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association. ''Circulation'', ''141'', E139–E596.</ref>. Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes<ref>Silva, G. S., & Nogueira, R. G. (2020). Endovascular treatment of acute ischemic stroke. ''CONTINUUM: Lifelong Learning in Neurology'', ''26''(2), 310–331.</ref>. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA)<ref>Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. ''Neurosurgery'', ''85''(suppl_1), S4–S8.</ref>. Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone<ref name=":0">McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In ''Scientific World Journal'' (Vol. 2019).</ref>. Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state<ref name=":0" />. However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit<ref name=":1">Goyal, M., Jadhav, A. P., Bonafe, A., Diener, H., Mendes Pereira, V., Levy, E., Baxter, B., Jovin, T., Jahan, R., & Menon, B. K. (2016). Analysis of workflow and time to treatment and the effects on outcome in endovascular treatment of acute ischemic stroke: results from the SWIFT PRIME randomized controlled trial. ''Radiology'', ''279''(3), 888–897.</ref>. Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred<ref name=":1" />. The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system<ref>Kunz, W. G., Hunink, M. G., Almekhlafi, M. A., Menon, B. K., Saver, J. L., Dippel, D. W. J., Majoie, C. B. L. M., Jovin, T. G., Davalos, A., & Bracard, S. (2020). Public health and cost consequences of time delays to thrombectomy for acute ischemic stroke. ''Neurology'', ''95''(18), e2465–e2475.</ref>. These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. ===Background=== Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain [20, 22]. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation [25, 49], resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==References== 64ttlp5yvalifmwz4us1f9nrg92zmua 563 562 2023-12-16T00:22:32Z KedarGrama 6 /* Background */ 563 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)<ref>Virani, S. S., Alonso, A., Benjamin, E. J., Bittencourt, M. S., Callaway, C. W., Carson, A. P., Chamberlain, A. M., Chang, A. R., Cheng, S., Delling, F. N., Djousse, L., Elkind, M. S. V., Ferguson, J. F., Fornage, M., Khan, S. S., Kissela, B. M., Knutson, K. L., Kwan, T. W., Lackland, D. T., ... Subcommittee, O. behalf of the A. H. A. C. on E. and P. S. C. and S. S. (2020). Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association. ''Circulation'', ''141'', E139–E596.</ref>. Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes<ref>Silva, G. S., & Nogueira, R. G. (2020). Endovascular treatment of acute ischemic stroke. ''CONTINUUM: Lifelong Learning in Neurology'', ''26''(2), 310–331.</ref>. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA)<ref>Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. ''Neurosurgery'', ''85''(suppl_1), S4–S8.</ref>. Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone<ref name=":0">McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In ''Scientific World Journal'' (Vol. 2019).</ref>. Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state<ref name=":0" />. However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit<ref name=":1">Goyal, M., Jadhav, A. P., Bonafe, A., Diener, H., Mendes Pereira, V., Levy, E., Baxter, B., Jovin, T., Jahan, R., & Menon, B. K. (2016). Analysis of workflow and time to treatment and the effects on outcome in endovascular treatment of acute ischemic stroke: results from the SWIFT PRIME randomized controlled trial. ''Radiology'', ''279''(3), 888–897.</ref>. Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred<ref name=":1" />. The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system<ref>Kunz, W. G., Hunink, M. G., Almekhlafi, M. A., Menon, B. K., Saver, J. L., Dippel, D. W. J., Majoie, C. B. L. M., Jovin, T. G., Davalos, A., & Bracard, S. (2020). Public health and cost consequences of time delays to thrombectomy for acute ischemic stroke. ''Neurology'', ''95''(18), e2465–e2475.</ref>. These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. ===Background=== Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain<ref>Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation (tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clin Neurophysiol. 2006 Apr;117(4):845-50</ref><ref>Hallett M: Transcranial magnetic stimulation and the human brain. Nature 2000; 406:147–150][Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation(tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clinical Neurophysiology. 2006 Apr;117(4):845-50</ref>. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation<ref>Huerta PT, Volpe BT: Transcranial magnetic stimulation, synaptic plasticity, and network oscillations. J Neuroeng Rehabil 2009]</ref><ref>Ogiue-Ikeda M, Kawato S, Ueno S: The effect of repetitive transcranial magnetic stimulation on long-term potentiation in rat hippocampus depends on stimulus intensity. Brain Res 2003; 993:222–226</ref>, resulting in changes in functional connectivity [19]. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation [37]. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.” [53].  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal [53].  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation [53]. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz) [7]. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa [1] # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==References== tpa1hmxqnwatywpbr5rk8cwq2sxodb3 564 563 2023-12-16T00:29:13Z KedarGrama 6 /* Background */ 564 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)<ref>Virani, S. S., Alonso, A., Benjamin, E. J., Bittencourt, M. S., Callaway, C. W., Carson, A. P., Chamberlain, A. M., Chang, A. R., Cheng, S., Delling, F. N., Djousse, L., Elkind, M. S. V., Ferguson, J. F., Fornage, M., Khan, S. S., Kissela, B. M., Knutson, K. L., Kwan, T. W., Lackland, D. T., ... Subcommittee, O. behalf of the A. H. A. C. on E. and P. S. C. and S. S. (2020). Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association. ''Circulation'', ''141'', E139–E596.</ref>. Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes<ref>Silva, G. S., & Nogueira, R. G. (2020). Endovascular treatment of acute ischemic stroke. ''CONTINUUM: Lifelong Learning in Neurology'', ''26''(2), 310–331.</ref>. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA)<ref>Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. ''Neurosurgery'', ''85''(suppl_1), S4–S8.</ref>. Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone<ref name=":0">McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In ''Scientific World Journal'' (Vol. 2019).</ref>. Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state<ref name=":0" />. However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit<ref name=":1">Goyal, M., Jadhav, A. P., Bonafe, A., Diener, H., Mendes Pereira, V., Levy, E., Baxter, B., Jovin, T., Jahan, R., & Menon, B. K. (2016). Analysis of workflow and time to treatment and the effects on outcome in endovascular treatment of acute ischemic stroke: results from the SWIFT PRIME randomized controlled trial. ''Radiology'', ''279''(3), 888–897.</ref>. Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred<ref name=":1" />. The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system<ref>Kunz, W. G., Hunink, M. G., Almekhlafi, M. A., Menon, B. K., Saver, J. L., Dippel, D. W. J., Majoie, C. B. L. M., Jovin, T. G., Davalos, A., & Bracard, S. (2020). Public health and cost consequences of time delays to thrombectomy for acute ischemic stroke. ''Neurology'', ''95''(18), e2465–e2475.</ref>. These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. ===Background=== Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain<ref>Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation (tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clin Neurophysiol. 2006 Apr;117(4):845-50</ref><ref>Hallett M: Transcranial magnetic stimulation and the human brain. Nature 2000; 406:147–150][Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation(tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clinical Neurophysiology. 2006 Apr;117(4):845-50</ref>. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation<ref>Huerta PT, Volpe BT: Transcranial magnetic stimulation, synaptic plasticity, and network oscillations. J Neuroeng Rehabil 2009]</ref><ref>Ogiue-Ikeda M, Kawato S, Ueno S: The effect of repetitive transcranial magnetic stimulation on long-term potentiation in rat hippocampus depends on stimulus intensity. Brain Res 2003; 993:222–226</ref>, resulting in changes in functional connectivity<ref>Fox MD, Buckner RL, White MP, et al: Efficacy of transcranial magnetic stimulation targets for depression is related to intrinsic functional connectivity with the subgenual cingulate. Biol Psychiatry 2012; 72:595–603].</ref>. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation<ref>Lucena MFG, Teixeira PEP, Bonin Pinto C, Fregni F. Top 100 cited noninvasive neuromodulation clinical trials. Expert Rev Med Devices. 2019 Jun;16(6):451-466</ref>. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation [13]. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus [59], amygdala [17], and thalamus [12]. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain [23]. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study [34]. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic [16, 17, 47, 60, 37]. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile [49]. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.”<ref name=":2">Sanguinetti JL, Hameroff S, Smith EE, et al. Transcranial Focused Ultrasound to the Right Prefrontal Cortex Improves Mood and Alters Functional Connectivity in Humans. Front Hum Neurosci. 2020;14:52.</ref>.  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal<ref name=":2" />.  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation<ref name=":2" />. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz)<ref>Blackmore, J., Shrivastava, S., Sallet, J., Butler, C. R., & Cleveland, R. O. (2019). Ultrasound neuromodulation: a review of results, mechanisms and safety. Ultrasound in medicine & biology, 45(7), 1509-1536.</ref>. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa<ref>AIUM. Statement on Biological Effects of Ultrasound in Vivo. <nowiki>https://www.aium.org/officialStatements/82</nowiki>. Last accessed February 21, 2022.</ref> # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==References== k17jr41x9d3g7r6sdkayccrkmxml03o 567 564 2023-12-16T00:37:19Z KedarGrama 6 /* Background */ 567 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)<ref>Virani, S. S., Alonso, A., Benjamin, E. J., Bittencourt, M. S., Callaway, C. W., Carson, A. P., Chamberlain, A. M., Chang, A. R., Cheng, S., Delling, F. N., Djousse, L., Elkind, M. S. V., Ferguson, J. F., Fornage, M., Khan, S. S., Kissela, B. M., Knutson, K. L., Kwan, T. W., Lackland, D. T., ... Subcommittee, O. behalf of the A. H. A. C. on E. and P. S. C. and S. S. (2020). Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association. ''Circulation'', ''141'', E139–E596.</ref>. Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes<ref>Silva, G. S., & Nogueira, R. G. (2020). Endovascular treatment of acute ischemic stroke. ''CONTINUUM: Lifelong Learning in Neurology'', ''26''(2), 310–331.</ref>. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA)<ref>Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. ''Neurosurgery'', ''85''(suppl_1), S4–S8.</ref>. Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone<ref name=":0">McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In ''Scientific World Journal'' (Vol. 2019).</ref>. Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state<ref name=":0" />. However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit<ref name=":1">Goyal, M., Jadhav, A. P., Bonafe, A., Diener, H., Mendes Pereira, V., Levy, E., Baxter, B., Jovin, T., Jahan, R., & Menon, B. K. (2016). Analysis of workflow and time to treatment and the effects on outcome in endovascular treatment of acute ischemic stroke: results from the SWIFT PRIME randomized controlled trial. ''Radiology'', ''279''(3), 888–897.</ref>. Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred<ref name=":1" />. The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system<ref>Kunz, W. G., Hunink, M. G., Almekhlafi, M. A., Menon, B. K., Saver, J. L., Dippel, D. W. J., Majoie, C. B. L. M., Jovin, T. G., Davalos, A., & Bracard, S. (2020). Public health and cost consequences of time delays to thrombectomy for acute ischemic stroke. ''Neurology'', ''95''(18), e2465–e2475.</ref>. These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. ===Background=== Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain<ref>Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation (tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clin Neurophysiol. 2006 Apr;117(4):845-50</ref><ref>Hallett M: Transcranial magnetic stimulation and the human brain. Nature 2000; 406:147–150][Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation(tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clinical Neurophysiology. 2006 Apr;117(4):845-50</ref>. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation<ref>Huerta PT, Volpe BT: Transcranial magnetic stimulation, synaptic plasticity, and network oscillations. J Neuroeng Rehabil 2009]</ref><ref>Ogiue-Ikeda M, Kawato S, Ueno S: The effect of repetitive transcranial magnetic stimulation on long-term potentiation in rat hippocampus depends on stimulus intensity. Brain Res 2003; 993:222–226</ref>, resulting in changes in functional connectivity<ref>Fox MD, Buckner RL, White MP, et al: Efficacy of transcranial magnetic stimulation targets for depression is related to intrinsic functional connectivity with the subgenual cingulate. Biol Psychiatry 2012; 72:595–603].</ref>. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation<ref>Lucena MFG, Teixeira PEP, Bonin Pinto C, Fregni F. Top 100 cited noninvasive neuromodulation clinical trials. Expert Rev Med Devices. 2019 Jun;16(6):451-466</ref>. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation<ref>di Biase L, Falato E, Di Lazzaro V. Transcranial Focused Ultrasound (tFUS) and Transcranial Unfocused Ultrasound (tUS) Neuromodulation: From Theoretical Principles to Stimulation Practices. Front Neurol 2019;10:549</ref>. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus<ref>Tufail Y, Matyushov A, Baldwin N, Tauchmann ML, Georges J, Yoshihiro A, et al. . Transcranial pulsed ultrasound stimulates intact brain circuits. Neuron. (2010) 66:681–94], amygdala [Folloni D, Verhagen L, Mars RB, Fouragnan E, Constans C, Aubry J-F, et al. . Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation. Neuron. (2019) 101:1109–16.e1105</ref>, amygdala<ref name=":3">Folloni D, Verhagen L, Mars RB, Fouragnan E, Constans C, Aubry J-F, et al. . Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation. Neuron. (2019) 101:1109–16.e1105</ref>, and thalamus<ref>Dallapiazza RF, Timbie KF, Holmberg S, Gatesman J, Lopes MB, Price RJ, et al. Non-invasive neuromodulation and thalamic mapping with low-intensity focused ultrasound. J Neurosurg.(2018) 128:875–84</ref>. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain<ref>Hameroff S, Trakas M, Duffield C, Annabi E, Gerace MB, Boyle P, et al. Transcranial ultrasound (TUS) effects on mental states: a pilot study. Brain Stimul. (2013) 6:409–15</ref>. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study<ref>Legon W, Sato TF, Opitz A, Mueller J, Barbour A, Williams A, et al. . Transcranial focused ultrasound modulates the activity of primary somatosensory cortex in humans. Nat Neurosci. (2014) 17:322–9</ref>. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic <ref name=":3" /><ref>Fini M, Tyler WJ. Transcranial focused ultrasound: a new tool for non-invasive neuromodulation. Int Rev Psychiatry. (2017) 29:168–77</ref><ref>Lucena MFG, Teixeira PEP, Bonin Pinto C, Fregni F. Top 100 cited noninvasive neuromodulation clinical trials. Expert Rev Med Devices. 2019 Jun;16(6):451-466</ref><ref>Munoz F, Aurup C, Konofagou EE, Ferrera VP. Modulation of brain function and behavior by focused ultrasound. Curr Behav Neurosci Rep. (2018) 5:153–64.</ref><ref>Tyler WJ, Lani SW, Hwang GM. Ultrasonic modulation of neural circuit activity. Curr Opinion Neurobiol. (2018) 50:222–31.</ref>. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile<ref>Ogiue-Ikeda M, Kawato S, Ueno S: The effect of repetitive transcranial magnetic stimulation on long-term potentiation in rat hippocampus depends on stimulus intensity. Brain Res 2003; 993:222–226</ref>. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.”<ref name=":2">Sanguinetti JL, Hameroff S, Smith EE, et al. Transcranial Focused Ultrasound to the Right Prefrontal Cortex Improves Mood and Alters Functional Connectivity in Humans. Front Hum Neurosci. 2020;14:52.</ref>.  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal<ref name=":2" />.  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation<ref name=":2" />. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz)<ref>Blackmore, J., Shrivastava, S., Sallet, J., Butler, C. R., & Cleveland, R. O. (2019). Ultrasound neuromodulation: a review of results, mechanisms and safety. Ultrasound in medicine & biology, 45(7), 1509-1536.</ref>. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer [7]. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa<ref>AIUM. Statement on Biological Effects of Ultrasound in Vivo. <nowiki>https://www.aium.org/officialStatements/82</nowiki>. Last accessed February 21, 2022.</ref> # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==References== 8f2x9grq0oohv5s97eqgdigoht5kgjl 568 567 2023-12-16T00:38:25Z KedarGrama 6 /* Communication With the FDA */ 568 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)<ref>Virani, S. S., Alonso, A., Benjamin, E. J., Bittencourt, M. S., Callaway, C. W., Carson, A. P., Chamberlain, A. M., Chang, A. R., Cheng, S., Delling, F. N., Djousse, L., Elkind, M. S. V., Ferguson, J. F., Fornage, M., Khan, S. S., Kissela, B. M., Knutson, K. L., Kwan, T. W., Lackland, D. T., ... Subcommittee, O. behalf of the A. H. A. C. on E. and P. S. C. and S. S. (2020). Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association. ''Circulation'', ''141'', E139–E596.</ref>. Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes<ref>Silva, G. S., & Nogueira, R. G. (2020). Endovascular treatment of acute ischemic stroke. ''CONTINUUM: Lifelong Learning in Neurology'', ''26''(2), 310–331.</ref>. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA)<ref>Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. ''Neurosurgery'', ''85''(suppl_1), S4–S8.</ref>. Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone<ref name=":0">McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In ''Scientific World Journal'' (Vol. 2019).</ref>. Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state<ref name=":0" />. However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit<ref name=":1">Goyal, M., Jadhav, A. P., Bonafe, A., Diener, H., Mendes Pereira, V., Levy, E., Baxter, B., Jovin, T., Jahan, R., & Menon, B. K. (2016). Analysis of workflow and time to treatment and the effects on outcome in endovascular treatment of acute ischemic stroke: results from the SWIFT PRIME randomized controlled trial. ''Radiology'', ''279''(3), 888–897.</ref>. Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred<ref name=":1" />. The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system<ref>Kunz, W. G., Hunink, M. G., Almekhlafi, M. A., Menon, B. K., Saver, J. L., Dippel, D. W. J., Majoie, C. B. L. M., Jovin, T. G., Davalos, A., & Bracard, S. (2020). Public health and cost consequences of time delays to thrombectomy for acute ischemic stroke. ''Neurology'', ''95''(18), e2465–e2475.</ref>. These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. ===Background=== Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain<ref>Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation (tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clin Neurophysiol. 2006 Apr;117(4):845-50</ref><ref>Hallett M: Transcranial magnetic stimulation and the human brain. Nature 2000; 406:147–150][Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation(tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clinical Neurophysiology. 2006 Apr;117(4):845-50</ref>. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation<ref>Huerta PT, Volpe BT: Transcranial magnetic stimulation, synaptic plasticity, and network oscillations. J Neuroeng Rehabil 2009]</ref><ref>Ogiue-Ikeda M, Kawato S, Ueno S: The effect of repetitive transcranial magnetic stimulation on long-term potentiation in rat hippocampus depends on stimulus intensity. Brain Res 2003; 993:222–226</ref>, resulting in changes in functional connectivity<ref>Fox MD, Buckner RL, White MP, et al: Efficacy of transcranial magnetic stimulation targets for depression is related to intrinsic functional connectivity with the subgenual cingulate. Biol Psychiatry 2012; 72:595–603].</ref>. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation<ref>Lucena MFG, Teixeira PEP, Bonin Pinto C, Fregni F. Top 100 cited noninvasive neuromodulation clinical trials. Expert Rev Med Devices. 2019 Jun;16(6):451-466</ref>. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation<ref>di Biase L, Falato E, Di Lazzaro V. Transcranial Focused Ultrasound (tFUS) and Transcranial Unfocused Ultrasound (tUS) Neuromodulation: From Theoretical Principles to Stimulation Practices. Front Neurol 2019;10:549</ref>. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus<ref>Tufail Y, Matyushov A, Baldwin N, Tauchmann ML, Georges J, Yoshihiro A, et al. . Transcranial pulsed ultrasound stimulates intact brain circuits. Neuron. (2010) 66:681–94], amygdala [Folloni D, Verhagen L, Mars RB, Fouragnan E, Constans C, Aubry J-F, et al. . Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation. Neuron. (2019) 101:1109–16.e1105</ref>, amygdala<ref name=":3">Folloni D, Verhagen L, Mars RB, Fouragnan E, Constans C, Aubry J-F, et al. . Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation. Neuron. (2019) 101:1109–16.e1105</ref>, and thalamus<ref>Dallapiazza RF, Timbie KF, Holmberg S, Gatesman J, Lopes MB, Price RJ, et al. Non-invasive neuromodulation and thalamic mapping with low-intensity focused ultrasound. J Neurosurg.(2018) 128:875–84</ref>. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain<ref>Hameroff S, Trakas M, Duffield C, Annabi E, Gerace MB, Boyle P, et al. Transcranial ultrasound (TUS) effects on mental states: a pilot study. Brain Stimul. (2013) 6:409–15</ref>. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study<ref>Legon W, Sato TF, Opitz A, Mueller J, Barbour A, Williams A, et al. . Transcranial focused ultrasound modulates the activity of primary somatosensory cortex in humans. Nat Neurosci. (2014) 17:322–9</ref>. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic <ref name=":3" /><ref>Fini M, Tyler WJ. Transcranial focused ultrasound: a new tool for non-invasive neuromodulation. Int Rev Psychiatry. (2017) 29:168–77</ref><ref>Lucena MFG, Teixeira PEP, Bonin Pinto C, Fregni F. Top 100 cited noninvasive neuromodulation clinical trials. Expert Rev Med Devices. 2019 Jun;16(6):451-466</ref><ref>Munoz F, Aurup C, Konofagou EE, Ferrera VP. Modulation of brain function and behavior by focused ultrasound. Curr Behav Neurosci Rep. (2018) 5:153–64.</ref><ref>Tyler WJ, Lani SW, Hwang GM. Ultrasonic modulation of neural circuit activity. Curr Opinion Neurobiol. (2018) 50:222–31.</ref>. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile<ref>Ogiue-Ikeda M, Kawato S, Ueno S: The effect of repetitive transcranial magnetic stimulation on long-term potentiation in rat hippocampus depends on stimulus intensity. Brain Res 2003; 993:222–226</ref>. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.”<ref name=":2">Sanguinetti JL, Hameroff S, Smith EE, et al. Transcranial Focused Ultrasound to the Right Prefrontal Cortex Improves Mood and Alters Functional Connectivity in Humans. Front Hum Neurosci. 2020;14:52.</ref>.  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal<ref name=":2" />.  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation<ref name=":2" />. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz)<ref>Blackmore, J., Shrivastava, S., Sallet, J., Butler, C. R., & Cleveland, R. O. (2019). Ultrasound neuromodulation: a review of results, mechanisms and safety. Ultrasound in medicine & biology, 45(7), 1509-1536.</ref>. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer<ref>Blackmore, J., Shrivastava, S., Sallet, J., Butler, C. R., & Cleveland, R. O. (2019). Ultrasound neuromodulation: a review of results, mechanisms and safety. Ultrasound in medicine & biology, 45(7), 1509-1536.</ref>. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa<ref>AIUM. Statement on Biological Effects of Ultrasound in Vivo. <nowiki>https://www.aium.org/officialStatements/82</nowiki>. Last accessed February 21, 2022.</ref> # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==References== jlw3qu1nlvpz5uadvuhkwnj1ecvgrqy 606 568 2023-12-18T22:39:10Z 50.227.118.138 606 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)<ref>Virani, S. S., Alonso, A., Benjamin, E. J., Bittencourt, M. S., Callaway, C. W., Carson, A. P., Chamberlain, A. M., Chang, A. R., Cheng, S., Delling, F. N., Djousse, L., Elkind, M. S. V., Ferguson, J. F., Fornage, M., Khan, S. S., Kissela, B. M., Knutson, K. L., Kwan, T. W., Lackland, D. T., ... Subcommittee, O. behalf of the A. H. A. C. on E. and P. S. C. and S. S. (2020). Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association. ''Circulation'', ''141'', E139–E596.</ref>. Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes<ref>Silva, G. S., & Nogueira, R. G. (2020). Endovascular treatment of acute ischemic stroke. ''CONTINUUM: Lifelong Learning in Neurology'', ''26''(2), 310–331.</ref>. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA)<ref>Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. ''Neurosurgery'', ''85''(suppl_1), S4–S8.</ref>. Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone<ref name=":0">McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In ''Scientific World Journal'' (Vol. 2019).</ref>. Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state<ref name=":0" />. However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit<ref name=":1">Goyal, M., Jadhav, A. P., Bonafe, A., Diener, H., Mendes Pereira, V., Levy, E., Baxter, B., Jovin, T., Jahan, R., & Menon, B. K. (2016). Analysis of workflow and time to treatment and the effects on outcome in endovascular treatment of acute ischemic stroke: results from the SWIFT PRIME randomized controlled trial. ''Radiology'', ''279''(3), 888–897.</ref>. Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred<ref name=":1" />. The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system<ref>Kunz, W. G., Hunink, M. G., Almekhlafi, M. A., Menon, B. K., Saver, J. L., Dippel, D. W. J., Majoie, C. B. L. M., Jovin, T. G., Davalos, A., & Bracard, S. (2020). Public health and cost consequences of time delays to thrombectomy for acute ischemic stroke. ''Neurology'', ''95''(18), e2465–e2475.</ref>. These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. ===Background=== Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain<ref>Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation (tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clin Neurophysiol. 2006 Apr;117(4):845-50</ref><ref>Hallett M: Transcranial magnetic stimulation and the human brain. Nature 2000; 406:147–150][Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation(tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clinical Neurophysiology. 2006 Apr;117(4):845-50</ref>. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation<ref>Huerta PT, Volpe BT: Transcranial magnetic stimulation, synaptic plasticity, and network oscillations. J Neuroeng Rehabil 2009]</ref><ref>Ogiue-Ikeda M, Kawato S, Ueno S: The effect of repetitive transcranial magnetic stimulation on long-term potentiation in rat hippocampus depends on stimulus intensity. Brain Res 2003; 993:222–226</ref>, resulting in changes in functional connectivity<ref>Fox MD, Buckner RL, White MP, et al: Efficacy of transcranial magnetic stimulation targets for depression is related to intrinsic functional connectivity with the subgenual cingulate. Biol Psychiatry 2012; 72:595–603].</ref>. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation<ref>Lucena MFG, Teixeira PEP, Bonin Pinto C, Fregni F. Top 100 cited noninvasive neuromodulation clinical trials. Expert Rev Med Devices. 2019 Jun;16(6):451-466</ref>. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation<ref>di Biase L, Falato E, Di Lazzaro V. Transcranial Focused Ultrasound (tFUS) and Transcranial Unfocused Ultrasound (tUS) Neuromodulation: From Theoretical Principles to Stimulation Practices. Front Neurol 2019;10:549</ref>. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus<ref>Tufail Y, Matyushov A, Baldwin N, Tauchmann ML, Georges J, Yoshihiro A, et al. . Transcranial pulsed ultrasound stimulates intact brain circuits. Neuron. (2010) 66:681–94], amygdala [Folloni D, Verhagen L, Mars RB, Fouragnan E, Constans C, Aubry J-F, et al. . Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation. Neuron. (2019) 101:1109–16.e1105</ref>, amygdala<ref name=":3">Folloni D, Verhagen L, Mars RB, Fouragnan E, Constans C, Aubry J-F, et al. . Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation. Neuron. (2019) 101:1109–16.e1105</ref>, and thalamus<ref>Dallapiazza RF, Timbie KF, Holmberg S, Gatesman J, Lopes MB, Price RJ, et al. Non-invasive neuromodulation and thalamic mapping with low-intensity focused ultrasound. J Neurosurg.(2018) 128:875–84</ref>. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain<ref>Hameroff S, Trakas M, Duffield C, Annabi E, Gerace MB, Boyle P, et al. Transcranial ultrasound (TUS) effects on mental states: a pilot study. Brain Stimul. (2013) 6:409–15</ref>. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study<ref>Legon W, Sato TF, Opitz A, Mueller J, Barbour A, Williams A, et al. . Transcranial focused ultrasound modulates the activity of primary somatosensory cortex in humans. Nat Neurosci. (2014) 17:322–9</ref>. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic <ref name=":3" /><ref>Fini M, Tyler WJ. Transcranial focused ultrasound: a new tool for non-invasive neuromodulation. Int Rev Psychiatry. (2017) 29:168–77</ref><ref>Lucena MFG, Teixeira PEP, Bonin Pinto C, Fregni F. Top 100 cited noninvasive neuromodulation clinical trials. Expert Rev Med Devices. 2019 Jun;16(6):451-466</ref><ref>Munoz F, Aurup C, Konofagou EE, Ferrera VP. Modulation of brain function and behavior by focused ultrasound. Curr Behav Neurosci Rep. (2018) 5:153–64.</ref><ref>Tyler WJ, Lani SW, Hwang GM. Ultrasonic modulation of neural circuit activity. Curr Opinion Neurobiol. (2018) 50:222–31.</ref>. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile<ref>Ogiue-Ikeda M, Kawato S, Ueno S: The effect of repetitive transcranial magnetic stimulation on long-term potentiation in rat hippocampus depends on stimulus intensity. Brain Res 2003; 993:222–226</ref>. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.”<ref name=":2">Sanguinetti JL, Hameroff S, Smith EE, et al. Transcranial Focused Ultrasound to the Right Prefrontal Cortex Improves Mood and Alters Functional Connectivity in Humans. Front Hum Neurosci. 2020;14:52.</ref>.  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal<ref name=":2" />.  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation<ref name=":2" />. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz)<ref>Blackmore, J., Shrivastava, S., Sallet, J., Butler, C. R., & Cleveland, R. O. (2019). Ultrasound neuromodulation: a review of results, mechanisms and safety. Ultrasound in medicine & biology, 45(7), 1509-1536.</ref>. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer<ref>Blackmore, J., Shrivastava, S., Sallet, J., Butler, C. R., & Cleveland, R. O. (2019). Ultrasound neuromodulation: a review of results, mechanisms and safety. Ultrasound in medicine & biology, 45(7), 1509-1536.</ref>. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa<ref>AIUM. Statement on Biological Effects of Ultrasound in Vivo. <nowiki>https://www.aium.org/officialStatements/82</nowiki>. Last accessed February 21, 2022.</ref> # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==Templates for Safety Data Collection and Safety Disclosures== We share a few templates that we use for safety disclosures and to collect safety data in studies * [https://github.com/OpenwaterHealth/opw_regulatory/blob/95c5b66ea78161b79a986df5e78fd55280db47fe/Safety%20Disclosure.docx Safety Disclosure] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/95c5b66ea78161b79a986df5e78fd55280db47fe/APPENDIX%20A%20-%20Safety%20Data%20Collection%20Template.docx Safety Data Collection Template] ==References== cgfqy52jnsbvnyw9mmgv7ntkjmu0dl9 636 606 2023-12-19T18:43:38Z Admin 1 Protected "[[Regulatory]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 606 wikitext text/x-wiki ==Introduction== The field of medical technology has witnessed remarkable advancements in recent years, with groundbreaking innovations reshaping the landscape of diagnostics and therapeutic interventions. This Wiki page is dedicated to providing transparent insight into the regulatory strategy, discussions, and pathway that Openwater has pursued to date with the United States Food and Drug Administration (FDA) for the Openwater LVO Stroke Alert, the Openwater Blood flow Headset, and Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System. Among these transformative technologies, Openwater has developed the Openwater LVO Stroke Alert. The Openwater LVO Stroke Alert is intended for use as an early notification system to identify and communicate blood flow data on patients with suspected Large Vessel Occlusion of the internal carotid artery or the proximal middle cerebral artery. The proposed device is indicated for use in pre-hospital settings to allow medical personnel to make better routing decisions for patients suspected of anterior LVO.  This technology along with the Openwater Blood Flow Headset have the potential to provide valuable blood flow data to medical professionals. Low intensity focused ultrasound represents another cutting-edge technology that harnesses the power of ultrasound waves at lower energy levels for therapeutic purposes. This technique has shown promise in diverse medical fields, such as targeted drug delivery, tissue regeneration, and neuromodulation. The Openwater Low Intensity Focused (LOFU) Ultrasound Therapy System is intended for safe neurostimulation of the prefrontal cortex. As we navigate the intricacies of optical blood flow and low intensity focused ultrasound technologies, this page serves as a valuable resource for professionals, researchers, and enthusiasts seeking a comprehensive understanding of the FDA's regulatory oversight in this dynamic intersection of medical science and technology. All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub rep All of the available documents pertaining to Openwater's regulatory work can be found on its GitHub [https://github.com/OpenwaterHealth/opw_regulatory repository]. ==Openwater LVO Stroke Alert== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a request for Breakthrough Device Designation as defined in 515(b) of the FD&C Act (21 U.S.C. 360e-3(b)) with the Food and Drug Administration (FDA). The initial request was submitted on September 10, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. The Openwater LVO Stroke Alert is intended to for use as an early notification system to aid in the pre-hospital assessment of acute stroke patients with suspected large vessel occlusion of the internal carotid artery or proximal middle cerebral artery in med of emergent transport to an endovascular capable center. ===Background=== According to the World Health Organization, every year over 15 million people suffer a stroke, 5 million of whom die and another 5 million are left permanently disabled (World, 2002). In the United States alone, there is one new stroke approximately every 30 seconds, 87% of which are acute ischemic strokes (AIS)<ref>Virani, S. S., Alonso, A., Benjamin, E. J., Bittencourt, M. S., Callaway, C. W., Carson, A. P., Chamberlain, A. M., Chang, A. R., Cheng, S., Delling, F. N., Djousse, L., Elkind, M. S. V., Ferguson, J. F., Fornage, M., Khan, S. S., Kissela, B. M., Knutson, K. L., Kwan, T. W., Lackland, D. T., ... Subcommittee, O. behalf of the A. H. A. C. on E. and P. S. C. and S. S. (2020). Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association. ''Circulation'', ''141'', E139–E596.</ref>. Due to the rapid loss of brain tissue during stroke, emergent diagnosis and treatment is critical for improving stroke outcomes<ref>Silva, G. S., & Nogueira, R. G. (2020). Endovascular treatment of acute ischemic stroke. ''CONTINUUM: Lifelong Learning in Neurology'', ''26''(2), 310–331.</ref>. Large vessel occlusions (LVOs) due to acute blockages of the proximal intracranial anterior and posterior circulation account for up to 46% of AIS and are considered refractory to intravenous tissue plasminogen activator (tPA)<ref>Rennert, R. C., Wali, A. R., Steinberg, J. A., Santiago-Dieppa, D. R., Olson, S. E., Pannell, J. S., & Khalessi, A. A. (2019). Epidemiology, natural history, and clinical presentation of large vessel ischemic stroke. ''Neurosurgery'', ''85''(suppl_1), S4–S8.</ref>. Emergent transport for endovascular therapy has now become the standard of care for anterior LVO resulting in AIS, with five randomized clinical trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT) demonstrating significant clinical improvements in both recanalization rates and clinical outcomes when comparing endovascular treatment to medical therapy alone<ref name=":0">McCarthy, D. J., Diaz, A., Sheinberg, D. L., Snelling, B., Luther, E. M., Chen, S. H., Yavagal, D. R., Peterson, E. C., & Starke, R. M. (2019). Long-term outcomes of mechanical thrombectomy for stroke: A meta-analysis. In ''Scientific World Journal'' (Vol. 2019).</ref>. Anterior LVO is defined as proximal occlusion of the internal carotid artery (ICA) or middle cerebral artery (MCA) at the first segment branching off the ICA, which is called the M1 segment. Importantly, thrombectomy is only indicated for ICA and M1 occlusions, and therefore, identification of the specific subtype of AIS being caused by anterior LVO from an ICA or M1 occlusion is of greatest importance for thrombectomy workflows. Despite these advances, long term outcomes after endovascular therapy demonstrate that >55% of LVO patients undergoing thrombectomy are left with a poor outcome of death or severely disabled dependent state<ref name=":0" />. However, if thrombectomy can be performed within 2.5 hours of LVO stroke onset, there is a >90% chance of a good neurological outcome, with minimal or no deficit<ref name=":1">Goyal, M., Jadhav, A. P., Bonafe, A., Diener, H., Mendes Pereira, V., Levy, E., Baxter, B., Jovin, T., Jahan, R., & Menon, B. K. (2016). Analysis of workflow and time to treatment and the effects on outcome in endovascular treatment of acute ischemic stroke: results from the SWIFT PRIME randomized controlled trial. ''Radiology'', ''279''(3), 888–897.</ref>. Clinical trials have demonstrated that the average time from hospital arrival to arterial access to begin endovascular treatment (“door to groin” time) can be improved by 50 minutes when there is a pre-hospital notification that the patient is being transferred<ref name=":1" />. The median loss in net monetary benefit of thrombectomy is calculated to be $1059 per minute, with every 10-minute reduction in average workflow time calculated to result in a $250 million in savings annually across the US healthcare system<ref>Kunz, W. G., Hunink, M. G., Almekhlafi, M. A., Menon, B. K., Saver, J. L., Dippel, D. W. J., Majoie, C. B. L. M., Jovin, T. G., Davalos, A., & Bracard, S. (2020). Public health and cost consequences of time delays to thrombectomy for acute ischemic stroke. ''Neurology'', ''95''(18), e2465–e2475.</ref>. These findings highlight the critical impact a pre-hospital LVO stroke alert can have on post-hospital arrival workflows, as with advance notification the receiving hospital teams and resources can be mobilized in preparation for the patient ahead of time, in parallel to transport. As such, generating pre-hospital LVO stroke alerts in and of themselves would have significant clinical benefit in reducing post-hospital arrival workflows by 50 minutes on average, which would amount to an estimated $1.25 billion in savings annually across the US healthcare system. Openwater proposed to introduce a noninvasive early notification system to identify and communicate blood flow data on patients suspected of large vessel occlusion of the internal carotid artery or proximal middle cerebral artery. This portable device is intended to operate in a prehospital emergency setting to generate pre-hospital LVO stroke alerts that can improve current arrival time workflows, enable advanced notification to receiving hospitals, facilitate new EMS routing protocols that allow for direct transfer of eligible patients to an endovascular capable center to improve the   LVO stroke workflow for times to re-perfusion and allow more patients to achieve good neurological outcomes than is currently possible with existing workflows. ===Communication With the FDA=== * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/1.%20FDA%20Breakthrough%20Designation%20Request%20for%20Openwater%20LVO%20Stroke%20Alert%20(Q211873)%20September%2010%2C%202021.pdf Openwater Request for Breakthrough Device Designation for LVO Stroke Alert] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/2.%20FDA%20Initial%20AIQR%20Letter%20for%20Q211873%20November%2012%2C%202021.pdf FDA Initial AIQR Letter for Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/3.%20Openwater%20Response%20to%20AIRQ%20Q211873%20November%202021.pdf Openwater Response to AIRQ Q211873] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/4.%20Openwater%20AIQR%20Q211873%20Deficiency%20Response%20with%20Clinical%20Data%20August%2015%2C%202023.pdf Openwater AIQR Q211873 Deficiency Response with Clinical Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/5.%20FDA%20AIQR%20for%20Q211873%20Deficiency%20Response%20October%204%2C%202023.pdf FDA AIQR for Q211873 Deficiency Response] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/6.%20Openwater%20Q211873-S001-AIRQ-Response%20with%20Q-Submission%20Requested%20Data.pdf Openwater Q211873-S001-AIRQ-Response with Q-Submission Requested Data] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/5a257ef2bab6e6720e3500e25906f5b763ee6b3f/bloodflow/7.%20Q211873%20FDA%20Breakthrough%20Designation%20Rejection%20Letter%20October%2024%2C%202023.pdf Q211873 FDA Breakthrough Designation Rejection Letter] ==Openwater Bloodflow Headset== These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on December 2, 2021. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. ===Background=== The Openwater Headset is a non-invasive device intended to monitor blood flow in the brain, specifically in the anterior portion in adults. The Openwater Headset is a platform that acquires this blood flow information by utilizing an Openwater proprietary technology that combines laser speckle imaging and diffuse correlation spectroscopy. The prospective clinical value of data from Openwater Headset has not been demonstrated in disease states. Openwater Headset should not be used as the sole basis for diagnosis or therapy. Openwater Headset consists of a head-mounted, wearable headset with built-in optical fibers for the delivery of low power laser light to the subject. CMOS (complementary metal oxide semiconductor) image sensors are utilized for the collection of light from the subject and a console with electronics to drive the headset and process the signal. The Openwater Headset provides two cerebral blood flow information, namely, * Blood Flow Index: an Openwater proprietary format, based in laser speckle contrast analysis, that provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi) * Blood Volume Index: an Openwater proprietary format, based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb) === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/Openwater%20FDA%20Pre-submission%20Request%20for%20Optical%20Blood%20Flow%20Headset%20(Q212497)%20December%202%2C%202021.pdf Openwater FDA Pre-submission Request for Optical Blood Flow Headset (Q212497)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Bloodflow/FDA%20Written%20Response%20and%20Feedback%20for%20Q212497.pdf FDA Written Response and Feedback for Q212497] == Openwater Low Intensity Focused Ultrasound Therapy System (LOFU) == These documents are the original versions submitted to and received from the FDA and are available as linked PDF’s. Openwater is committed to transparency and welcomes your feedback and questions regarding our regulatory process and submissions. Openwater submitted a Q-Submission (Q-Sub) request with the Food and Drug Administration (FDA) to discuss regulatory pathway and strategy, performance testing, and specific feedback from the FDA on this particular device. The initial request was submitted on March 7, 2022. The information contained in these files details the information submitted to the FDA, feedback received from the FDA, and clinical data submitted to the FDA. Please read the entire FDA Q-Submission for detailed information and works cited. ===Background=== Transcranial Magnetic Stimulation (TMS) and Transcranial Direct Current Stimulation (tDCS)  are FDA-cleared technologies that use either electromagnetic induction or direct electrical current, respectively, to accomplish noninvasive activation of targeted regions within the brain<ref>Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation (tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clin Neurophysiol. 2006 Apr;117(4):845-50</ref><ref>Hallett M: Transcranial magnetic stimulation and the human brain. Nature 2000; 406:147–150][Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation(tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clinical Neurophysiology. 2006 Apr;117(4):845-50</ref>. TMS treatments employ a magnetic field generator positioned at specific areas on the surface of a patient’s head in order to deliver focused magnetic fields to desired brain regions. These magnetic fields pass unimpeded through the skin and skull to induce directed electric currents in specific neurons of the targeted brain region. Exploiting the fact that neurons are electrochemical cells, this electric current is powerful enough to induce action potentials that result in neurotransmitter release.  Repeated high-frequency excitation of the same brain region results in strengthening of synapses through a process known as long-term potentiation<ref>Huerta PT, Volpe BT: Transcranial magnetic stimulation, synaptic plasticity, and network oscillations. J Neuroeng Rehabil 2009]</ref><ref>Ogiue-Ikeda M, Kawato S, Ueno S: The effect of repetitive transcranial magnetic stimulation on long-term potentiation in rat hippocampus depends on stimulus intensity. Brain Res 2003; 993:222–226</ref>, resulting in changes in functional connectivity<ref>Fox MD, Buckner RL, White MP, et al: Efficacy of transcranial magnetic stimulation targets for depression is related to intrinsic functional connectivity with the subgenual cingulate. Biol Psychiatry 2012; 72:595–603].</ref>. Both tDCS and TMS have become important clinical tools in noninvasive brain stimulation<ref>Lucena MFG, Teixeira PEP, Bonin Pinto C, Fregni F. Top 100 cited noninvasive neuromodulation clinical trials. Expert Rev Med Devices. 2019 Jun;16(6):451-466</ref>. Transcranial focused ultrasound stimulation (tFUS) is an emerging technique for non-invasive neurostimulation with improved spatial resolution and targeting compared to magnetic or electric non-invasive brain stimulation<ref>di Biase L, Falato E, Di Lazzaro V. Transcranial Focused Ultrasound (tFUS) and Transcranial Unfocused Ultrasound (tUS) Neuromodulation: From Theoretical Principles to Stimulation Practices. Front Neurol 2019;10:549</ref>. A number of studies have been performed using tFUS for non-invasive neurostimulation of targeted brain regions, including of the primary motor cortex and hippocampus<ref>Tufail Y, Matyushov A, Baldwin N, Tauchmann ML, Georges J, Yoshihiro A, et al. . Transcranial pulsed ultrasound stimulates intact brain circuits. Neuron. (2010) 66:681–94], amygdala [Folloni D, Verhagen L, Mars RB, Fouragnan E, Constans C, Aubry J-F, et al. . Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation. Neuron. (2019) 101:1109–16.e1105</ref>, amygdala<ref name=":3">Folloni D, Verhagen L, Mars RB, Fouragnan E, Constans C, Aubry J-F, et al. . Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation. Neuron. (2019) 101:1109–16.e1105</ref>, and thalamus<ref>Dallapiazza RF, Timbie KF, Holmberg S, Gatesman J, Lopes MB, Price RJ, et al. Non-invasive neuromodulation and thalamic mapping with low-intensity focused ultrasound. J Neurosurg.(2018) 128:875–84</ref>. The first human transcranial application of focused ultrasound neuromodulation involved stimulation of the frontal cortex applied on 31 patients affected by chronic pain<ref>Hameroff S, Trakas M, Duffield C, Annabi E, Gerace MB, Boyle P, et al. Transcranial ultrasound (TUS) effects on mental states: a pilot study. Brain Stimul. (2013) 6:409–15</ref>. Subsequent use of the tFUS technique was described targeting the primary somatosensory cortex of healthy volunteers, in a within-patients, sham-controlled study<ref>Legon W, Sato TF, Opitz A, Mueller J, Barbour A, Williams A, et al. . Transcranial focused ultrasound modulates the activity of primary somatosensory cortex in humans. Nat Neurosci. (2014) 17:322–9</ref>. There has been significantly increased interest in the clinical community on the applications of tFUS on neuromodulation, with a number of recent reviews published in order to summarize the state of the art on this topic <ref name=":3" /><ref>Fini M, Tyler WJ. Transcranial focused ultrasound: a new tool for non-invasive neuromodulation. Int Rev Psychiatry. (2017) 29:168–77</ref><ref>Lucena MFG, Teixeira PEP, Bonin Pinto C, Fregni F. Top 100 cited noninvasive neuromodulation clinical trials. Expert Rev Med Devices. 2019 Jun;16(6):451-466</ref><ref>Munoz F, Aurup C, Konofagou EE, Ferrera VP. Modulation of brain function and behavior by focused ultrasound. Curr Behav Neurosci Rep. (2018) 5:153–64.</ref><ref>Tyler WJ, Lani SW, Hwang GM. Ultrasonic modulation of neural circuit activity. Curr Opinion Neurobiol. (2018) 50:222–31.</ref>. In particular, systematic meta-analyses on tFUS safety across 33 studies performed in both humans and animals has demonstrated a favorable safety profile<ref>Ogiue-Ikeda M, Kawato S, Ueno S: The effect of repetitive transcranial magnetic stimulation on long-term potentiation in rat hippocampus depends on stimulus intensity. Brain Res 2003; 993:222–226</ref>. Clinical interest in the availability of tFUS as a clinical research tool is based in the fact that “tFUS offers several advantages over electromagnetic methods including high spatial resolution and the ability to reach deep brain targets.”<ref name=":2">Sanguinetti JL, Hameroff S, Smith EE, et al. Transcranial Focused Ultrasound to the Right Prefrontal Cortex Improves Mood and Alters Functional Connectivity in Humans. Front Hum Neurosci. 2020;14:52.</ref>.  Early results of measured clinical effects of tFUS in modulating mood and functional connectivity by targeting the right inferior frontal gyrus (rIFG), an area implicated in mood and emotional regulation, have shown promising signal<ref name=":2" />.  Briefly, in a randomized, placebo-controlled, double-blind study, participants who received tFUS reported an overall increase in Global Affect, an aggregate score from the Visual Analog Mood Scales, consistent with a positive shift as also objectively measured via decrease in resting-state functional connectivity (FC) on functional magnetic resonance imaging (fMRI) within resting state networks related to emotion and mood regulation<ref name=":2" />. These results support tFUS as a safe and effective means of noninvasive transcranial brain stimulation that can be used to modulate mood and emotional regulation networks in the prefrontal cortex. Openwater is developing the Openwater LOFU Therapy System, which is a low intensity transcranial focused ultrasound system intended to be used safely, in clinical research, for neurostimulation that targets the prefrontal cortex. This presubmission focuses on describing this device, proposing a regulatory approach to market this device, and obtaining FDA feedback to ensure alignment with Agency expectations for a future premarket submission. The efficacy of neurostimulation using LOFU or the subject device is not in scope of this presubmission. The Openwater LOFU Therapy System (“OLTS”) is a low intensity focused ultrasound (LOFU) system intended for safe neurostimulation of the prefrontal cortex. One of the key uses of this system is to support clinical research and exploration of indications that can benefit from LOFU neurostimulation. Focused ultrasound (FUS) is a way of non-invasively delivering energy to specific points, or foci, in the form of an acoustic pressure wave. The acoustic waves can be focused to a particular location with a spatial resolution on the order of the wavelength of the driving frequency (approximately 3 mm at 0.5 MHz)<ref>Blackmore, J., Shrivastava, S., Sallet, J., Butler, C. R., & Cleveland, R. O. (2019). Ultrasound neuromodulation: a review of results, mechanisms and safety. Ultrasound in medicine & biology, 45(7), 1509-1536.</ref>. The point of focus is achieved using an ultrasound array, where each element of the array is driven in a specific sequence to ensure that a focal spot can be formed through constructive interference of the ultrasound waves. The focal spot can be formed at depth within the tissue without affecting cells along the propagation path closer to the transducer<ref>Blackmore, J., Shrivastava, S., Sallet, J., Butler, C. R., & Cleveland, R. O. (2019). Ultrasound neuromodulation: a review of results, mechanisms and safety. Ultrasound in medicine & biology, 45(7), 1509-1536.</ref>. In the case of LOFU used by OLTS, energy levels used for FUS are set to low intensity levels that do not cause thermal or mechanical damage to the target tissue and to any tissue between the device and the target (i.e., within diagnostic limits of MI < 1.9, and I_SPTA < 720mW/cm²). The device operates in two steps and safety is incorporated into both steps: # ''Treatment Planning'': in this step, the clinician uses the device user interface to provide the treatment volume to be stimulated as well as the desired stimulation parameters. The OLTS treatment planning software computes the optimal and safe patient specific simulation parameters and treatment plan. Treatment planning includes the computation of ultrasound mechanical index (MI) and cranial bone thermal index (TIC) to ensure that the ultrasound pressure and intensity levels are kept far below those that may cause tissue damage. The treatment plan also includes the sequencing parameters needed to drive each element in the OLTS’s ultrasound array to achieve the foci needed to cover the treatment volume. The software limits the device’s maximum peak-rarefactional pressure to be less than 1.5MPa for all driving frequencies offered by OLTS, which is well below the theoretical lower threshold for bubble nucleation or inertial cavitation of 3.9MPa<ref>AIUM. Statement on Biological Effects of Ultrasound in Vivo. <nowiki>https://www.aium.org/officialStatements/82</nowiki>. Last accessed February 21, 2022.</ref> # The treatment planning is discussed in detail in Section 3.3.1 of the Q-Submission. # ''Treatment Delivery'': in this step, the OLTS executes the treatment plan to deliver the neurostimulation. During the delivery, the OLTS monitors treatment safety control parameters such as ultrasound hardware output power, stimulation time, and surface temperature at the site of the ultrasound array to ensure safe operation. === Communication With the FDA === * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Pre-Submission%20Request%20for%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf Pre-Submission Request for Openwater LOFU Therapy System] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/FDA%20Written%20Feedback%20on%20Openwater%20LOFU%20Therapy%20System%20(Q220469).pdf FDA Written Feedback on Openwater LOFU Therapy System (Q220469)] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/9e151dde43bbd40daee82584ea5b3a016b6c7f37/Ultrasound/Openwater%20Response%20to%20FDA%20Written%20Feedback%20(Q220469).pdf Openwater Response to FDA Written Feedback (Q220469)] ==Templates for Safety Data Collection and Safety Disclosures== We share a few templates that we use for safety disclosures and to collect safety data in studies * [https://github.com/OpenwaterHealth/opw_regulatory/blob/95c5b66ea78161b79a986df5e78fd55280db47fe/Safety%20Disclosure.docx Safety Disclosure] * [https://github.com/OpenwaterHealth/opw_regulatory/blob/95c5b66ea78161b79a986df5e78fd55280db47fe/APPENDIX%20A%20-%20Safety%20Data%20Collection%20Template.docx Safety Data Collection Template] ==References== cgfqy52jnsbvnyw9mmgv7ntkjmu0dl9 0 29 93 2023-12-13T02:13:20Z 24.92.36.30 Created page with "== '''Gen 1 Stroke Detection Device''' == === '''Overview''' === The Gen 1 Stroke Detection Device is a pioneering medical instrument designed for the noninvasive measurement of blood flow in the human brain. This technology represents a significant advancement in stroke detection and monitoring, offering real-time, accurate assessment of cerebral blood circulation. === '''Principle''' === The device operates on the principle of Infrared laser speckle contrast analysis..." 93 wikitext text/x-wiki == '''Gen 1 Stroke Detection Device''' == === '''Overview''' === The Gen 1 Stroke Detection Device is a pioneering medical instrument designed for the noninvasive measurement of blood flow in the human brain. This technology represents a significant advancement in stroke detection and monitoring, offering real-time, accurate assessment of cerebral blood circulation. === '''Principle''' === The device operates on the principle of Infrared laser speckle contrast analysis with our novel system described in this repository.  Our system provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi).  In addition, we provide a measurement of Blood Volume Index based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb). We have optimized measurement of the sensitivity to blood flow changes within the cranial vessels and its ability to provide real-time feedback. === '''Components''' === Components that make up the system: # Source Module: Incorporates a high-precision Moglabs laser, which emits [specific wavelength] light. This module includes mirrors and an Acousto-Optic Modulator (AOM) for directing and modulating the laser beam. # Detector Module: Responsible for capturing the light reflected or scattered from the cranial tissues. This module helps in quantifying the blood flow based on changes in the properties of the detected light. # Octopus PCB: A custom Printed Circuit Board that integrates the hardware components and facilitates signal processing. # Wand: A handheld component for directing the laser towards the patient’s head. It includes safety interlocks and user feedback mechanisms. === '''Software Architecture''' === The device software uses a combination of Python and C++ for user interface and data processing. The software operates in three modes: # Align Mode: For initial setup and calibration. # Scan Mode: For conducting measurements. # Backup Mode: For data storage and retrieval. === '''Safety Features''' === Safety is paramount in the design of the Gen 1 Stroke Detection Device. It includes: * Laser Safety Board to regulate laser emission. * Emergency switches and interlocks for immediate cessation of operation if needed. * Compliance with relevant medical and laser safety standards. === '''Operation''' === The operation of the Gen 1 Stroke Detection Device involves the following steps: # Alignment: The device is aligned with the area of interest on the patient’s head using the wand. # Scanning: The source module emits a laser beam, which penetrates the skull and interacts with the brain tissue. The scattered light is captured by the detector module. # Data Processing: The captured light is processed to derive blood flow information, considering the Doppler shift or changes in light properties due to blood movement. # Feedback: The results are displayed in real-time, providing immediate insights into the cerebral blood flow dynamics. === '''Clinical Application''' === The device is intended for use in clinical settings to aid in the early detection and monitoring of stroke. It can also be employed in research studies to understand cerebral blood flow patterns in various neurological conditions. 0c3i322deruef82jukxkiofzpbq4deo 637 93 2023-12-19T18:43:48Z Admin 1 Protected "[[Theory of Operation]]" ([Edit=Allow only administrators] (indefinite) [Move=Allow only administrators] (indefinite)) 93 wikitext text/x-wiki == '''Gen 1 Stroke Detection Device''' == === '''Overview''' === The Gen 1 Stroke Detection Device is a pioneering medical instrument designed for the noninvasive measurement of blood flow in the human brain. This technology represents a significant advancement in stroke detection and monitoring, offering real-time, accurate assessment of cerebral blood circulation. === '''Principle''' === The device operates on the principle of Infrared laser speckle contrast analysis with our novel system described in this repository.  Our system provides a measure of the flow rate of blood in the underlying tissue below the sensor, similar to other transcranial optical devices that provide measurements of relative cerebral blood flow index (rCBFi).  In addition, we provide a measurement of Blood Volume Index based on measurements of the concentration of absorbing chromophores within the underlying tissue below the sensor, and similar to other transcranial optical devices that provide measurements of relative total tissue hemoglobin concentration (rTHb). We have optimized measurement of the sensitivity to blood flow changes within the cranial vessels and its ability to provide real-time feedback. === '''Components''' === Components that make up the system: # Source Module: Incorporates a high-precision Moglabs laser, which emits [specific wavelength] light. This module includes mirrors and an Acousto-Optic Modulator (AOM) for directing and modulating the laser beam. # Detector Module: Responsible for capturing the light reflected or scattered from the cranial tissues. This module helps in quantifying the blood flow based on changes in the properties of the detected light. # Octopus PCB: A custom Printed Circuit Board that integrates the hardware components and facilitates signal processing. # Wand: A handheld component for directing the laser towards the patient’s head. It includes safety interlocks and user feedback mechanisms. === '''Software Architecture''' === The device software uses a combination of Python and C++ for user interface and data processing. The software operates in three modes: # Align Mode: For initial setup and calibration. # Scan Mode: For conducting measurements. # Backup Mode: For data storage and retrieval. === '''Safety Features''' === Safety is paramount in the design of the Gen 1 Stroke Detection Device. It includes: * Laser Safety Board to regulate laser emission. * Emergency switches and interlocks for immediate cessation of operation if needed. * Compliance with relevant medical and laser safety standards. === '''Operation''' === The operation of the Gen 1 Stroke Detection Device involves the following steps: # Alignment: The device is aligned with the area of interest on the patient’s head using the wand. # Scanning: The source module emits a laser beam, which penetrates the skull and interacts with the brain tissue. The scattered light is captured by the detector module. # Data Processing: The captured light is processed to derive blood flow information, considering the Doppler shift or changes in light properties due to blood movement. # Feedback: The results are displayed in real-time, providing immediate insights into the cerebral blood flow dynamics. === '''Clinical Application''' === The device is intended for use in clinical settings to aid in the early detection and monitoring of stroke. It can also be employed in research studies to understand cerebral blood flow patterns in various neurological conditions. 0c3i322deruef82jukxkiofzpbq4deo 0 127 655 2023-12-20T22:36:41Z KedarGrama 6 KedarGrama moved page [[USTX Board]] to [[Openwater Ultrasound Transmit Module (USTX)]]: Descriptive title that Chris Bawiec demanded 655 wikitext text/x-wiki #REDIRECT [[Openwater Ultrasound Transmit Module (USTX)]] i0v597bvrez0zrwkua0enu7olxon1ud 0 33 122 2023-12-13T03:00:48Z Gvigelet 4 Gvigelet moved page [[Whitepaper]] to [[Openwater Stroke Diagnosis Technology]] 122 wikitext text/x-wiki #REDIRECT [[Openwater Stroke Diagnosis Technology]] r6aswjq437tx3ixls4v7uu1ifp85qsz